Well guess what this does?

Quote:

---

Originally posted by bigtoe
You don't want to know....High thats all I say.

---

4volts? 4.2, 4.5, 4.8, 5 volts???


or is it more fun for ocz employees if people fry their ram trying to find the max and have to buy new ram? Just give us an idea of what kind of voltages = quick death. these devices WILL be moddified for higher voltages and it would be great if ocz would help it's users out.

Most people consider above 3volts high. I run 3.4volts 24/7 so i consider 3.8volts high. its all relative.

Quote:

---

Originally posted by bigtoe
ns_ripper

First off no insult taken but im not an A64 tester, i have been intel for 2 yrs.This doesn't mean i can't test it, it just means i am a little behind to what i know about 875 and 865 boards etc.

I know exactly how the voltages effect ram and i know exactly how it works..the ONLY thing im personnally unsure about is VDD or VDDQ passing back to the memory controller on A64.

Some graphics cards work in a VERY similar way to the booster and have seperate plugable PSU on them supplying the Vdimm and vcore to the card.

With regards to Intel boards with seperate vreg for vdimm etc I see no issue.With the booster set at 2.5v and the board set at 2.8V the booster will report the bios set vdimm and apply nothing. If the booster is set at 2.8 and the board 2.5 the board will have 2.8Vdimm at the mosfets etc as if the bios has set 2.8V.

I see this no different to guys altering the 3.3vline to 3.7V with onboard vdimm mods and pushing 3.6vdimm thru the onboard mosfets etc, infact using a booster will actually create less stress on the board as the mosfets will not be delivering the current ...the booster will.

Regarding Intel boards we have actually pushed EB to the point it fried the IC's with absolutely NO DAMAGE to the motherboard what so ever....so we do test quite thoroughly ;)
Now...i do recognise A64 may be an issue.The modules are not on sale yet so we have time to tweak if needed.

So to conclude...have you posted this same info to Abit, MSI, Epox etc and all the others that allow vdimm of 3.2v or higher ???

---

There is no difference between someone modding in is board and doing this. Only you need to solder, and leave fysical evidence behind.

This is much easier to do, so every noob, who doesn't know what he is doing, can do it. Without knowing the risks involved. So a big disclaimer wouldn't be misplaced. It can protect you and ocs from upset noobs wich are willing to take legal proceedings because they didn't know the risks, ... So it's also for your own safety.

Many boards allow a certain level of overvolting. I assume (hope) that they use components wich can take that ammount of overvolting. They also have the exact chipset specifications, so they also know how much the memory controller can take. But as you know for yourself many A64's died in the past while a high VDIMM was used.

Shuttle boards seem pretty sucky at this. I visit Sudhian forum's pretty often for SSF stuff. And there is a certain motherboard wich simpely dies when you change the VDIMM. So shuttle made a new bios wich doesn't allow to change it anymore. Go figure ... According to shuttle, overvolting any part of the system, even from within the bios, voids it warrantee. Those are not my words, but those are shuttle's words. I will dig up that thread for you. So you can read it for yourself.

Don't get me wrong bigtoe, I realy appreciate what you guy's are doing to please the oc community. But be carfully. I Also try to help as many people as I can with computer related stuff, but people can become very unthankfully when something doesn't go as they planned.

Link
btw, I'm lodewj overthere.

Quote:

---

Originally posted by zabomb4163
4volts? 4.2, 4.5, 4.8, 5 volts???


or is it more fun for ocz employees if people fry their ram trying to find the max and have to buy new ram? Just give us an idea of what kind of voltages = quick death. these devices WILL be moddified for higher voltages and it would be great if ocz would help it's users out.

Most people consider above 3volts high. I run 3.4volts 24/7 so i consider 3.8volts high. its all relative.

---

The dimm won't go high enough for you to replicate what we did...so what is the point of me telling you what voltage we used?

Quote:

---

Originally posted by ns_ripper
There is no difference between someone modding in is board and doing this. Only you need to solder, and leave fysical evidence behind.

This is much easier to do, so every noob, who doesn't know what he is doing, can do it. Without knowing the risks involved. So a big disclaimer wouldn't be misplaced. It can protect you and ocs from upset noobs wich are willing to take legal proceedings because they didn't know the risks, ... So it's also for your own safety.

Many boards allow a certain level of overvolting. I assume (hope) that they use components wich can take that ammount of overvolting. They also have the exact chipset specifications, so they also know how much the memory controller can take. But as you know for yourself many A64's died in the past while a high VDIMM was used.

Shuttle boards seem pretty sucky at this. I visit Sudhian forum's pretty often for SSF stuff. And there is a certain motherboard wich simpely dies when you change the VDIMM. So shuttle made a new bios wich doesn't allow to change it anymore. Go figure ... According to shuttle, overvolting any part of the system, even from within the bios, voids it warrantee. Those are not my words, but those are shuttle's words. I will dig up that thread for you. So you can read it for yourself.

Don't get me wrong bigtoe, I realy appreciate what you guy's are doing to please the oc community. But be carfully. I Also try to help as many people as I can with computer related stuff, but people can become very unthankfully when something doesn't go as they planned.

Link
btw, I'm lodewj overthere.

---

I think you will find we will advise caution if overvolting with the dimm, we already do above 3V using bios or onboard vdimm mods.

You also failed to recognise the dimm conditions power also, you may find you can clock further at 2.7v using the dimm rather than setting 2.7 in bios due to the very clean power delivered from the dimm to the ram...so you may NOT have to ram 3.9V to get the clock you want.

And finally, we leaked the dimm info here as Xtreme members are well aware of the risks involved with ram overvolting. Many have been pushing much higher Vdimm than i ever have and have even professed to me they is no issue when i have tried to advise caution.

hence zabomb4163's question, how much vdimm did it take to fry the EB's?

Quote:

---

Originally posted by enzoR
hence zabomb4163's question, how much vdimm did it take to fry the EB's?

---

\

My guess is 4.5-5V. I've run ram at 4.2V for more than a few hours with nothing more than a fan and they are still alive. But if Tony doesn't want to say they he doesn't have to :toast:

Steve

Originally posted by bigtoe

Quote:

---

I think you will find we will advise caution if overvolting with the dimm, we already do above 3V using bios or onboard vdimm mods.

---

logically that voids the warrantee of the components wich you overvolt.
But as you can read in the thread I just posted, that shuttle said; running the vddr out of specs from within the bios, voids there motherboards warrantee. That is pretty unlogical, because they allow us too. But that's what shuttle said if you read that thread. I also reacted in that thread, that something like that is unlogical. But who am I ...

Quote:

---

You also failed to recognise the dimm conditions power also, you may find you can clock further at 2.7v using the dimm rather than setting 2.7 in bios due to the very clean power delivered from the dimm to the ram...so you may NOT have to ram 3.9V to get the clock you want.

---

I don't fail to recognise such a thing. It's not worth mentioning it because it's off the topic of potential dangers. I only posted my reacation because you asked this:

Quote:

---

If you know of any board that actually links the memory controller voltage to the vdimm please post where i may check it out.

---

and I answered that question.

The powercondition is a pretty logical thing also. A motherboard is designed to operate within it's specs. The oc part is just something wich is added to satisfy a small ammount of people. the enthousiasts. So manufacturors don't use stabalizers wich can provide a lot more voltage and current. This would be more costely, and maybe 1/10th of the people who buy such a board would see that as an advantage.

most boards use the 3.3V line to derive the ddr voltage from. When you use a stabaliser, it has a point of most effectiveness. You always have to feed some volts more. how much for the ideal efficiency is told in it's datasheet. when you go over that point, the voltage will come closer to the voltage fed to it. this results in an unstable voltage. going up and down. when you have much and large memory usage at a time, many capacitors are being charged/uncharched. so the current also rises at that moment. wich will cause some more jumpy voltages.

the thing you do is: take a line of the atx header: feed it into a more powerfull stableliser. the line is bigger, so there is less resistance, so you won't loose on the line. --> Resistance = Length * section * materialResistance. so a trace has more resistance. while that thick wire hasn't. so that's a bottleneck less.

and the stabalisers are more powerfull.
So --> Powerclean Technologyâ„¢ --> roflol. you don't filter anything. you just use a lead wich has less resistance than a trace on a mobo, and you use more powerfull stablelisers.

by the way, a clean powersignal can't be mesured with a simple voltmeter. It has to be mesured with a scope. When you operate memory at 200Mhz, that means a refresh of 200milion times.
your device has only one decimal and a refresh of 3 seconds.

So how can you possible represent clean power with that? if you would show some scope graphs then we could talk about clean.

When you want to make clean power, you have to use capacitor with a very low ESR, rated for high ripple current.

The module is basically a dimm pcb with 2 led blocks, a maxim programmable microcontroller with a oscilator and a pll, ... and 3 voltage stablelisers.

Quote:

---

And finally, we leaked the dimm info here as Xtreme members are well aware of the risks involved with ram overvolting. Many have been pushing much higher Vdimm than i ever have and have even professed to me they is no issue when i have tried to advise caution.

---

excuse me for the many spelling mistakes, but I'm still trying to study my exam for tomorrow, so it's time critical :)
I didn't want to sound rude, but when you say, I fail to ...
I don't have to prove myself.
You made a comercial product and said this: If you know of any board that actually links the memory controller voltage to the vdimm please post where i may check it out.

proving that you don't know what you are messing with.
So I get a little upset when someone who doesn't know what he is dealing with, sais that I fail to ... :rolleyes:

by the way, no hard feelings :toast:

ns_ripper

I have just spent 14hrs trying to link HT voltage to VDDQ/VDD on the ram so please back off and give me a break.

Im on the VNF3-250 and i have found Vchipset VHT (hypertransport) and ALL Memory voltages. The dimm does NOT effect HT voltage on this board, nor does it effect Vchipset.

I have just maxed my old beta 3500EB with 2.95V at 260fsb 3-3-2-9 4xHT with 9x260 and been looping 3dmark for 4 hrs without issue.

If you think you have a point to make show me where HT voltage is linked to VDD/VDDQ so that I may test it...if you can't please let the tweakers tweak and the ones who are NOT so sure (like you) please avoid this product and run things at stock.

Quote:

---

logically that voids the warrantee of the components wich you overvolt.
But as you can read in the thread I just posted, that shuttle said; running the vddr out of specs from within the bios, voids there motherboards warrantee.

---

LOL..I have the Vdimm mod for the 875G2 Xpc direct from shuttle...3.2Vdimm max.

Xtreme is not the place to be preaching about memory voltage as i know of next to no one on this forum who doesn't mod, tweak or generally push things to the limit. I tried a while ago to preach the same as you and all it got me was a thread of mild flames and "well its what we do on here"

Quote:

---

Originally posted by enzoR
hence zabomb4163's question, how much vdimm did it take to fry the EB's?

---

If I quote what voltage we took you guys will then consider that the max and run slightly lower.

by not saying im keeping you guys cautious.

keep up the great work bigtoe...........

Quote:

---

Originally posted by bigtoe
If I quote what voltage we took you guys will then consider that the max and run slightly lower.

by not saying im keeping you guys cautious.

---

:P

bigtoe, you should check sis chipset based boards. afaik at least some of the sis boards have very weird voltage circuits. on the K7S8XE (?) vdimm,vagp and vdd(chipset) were all related, and to get high vdimm you had to do an annoying mod to keep the vagp below 1.8v. well there are not many people with sis boards anyways, but id suggest you to put it on the NOT SUPPORTED list before you have tested a board.

Hi guys, before the Vdimm Booster is released we will have a list available showing the results of testing with the product. We have not found any NF3 boards so far that aren't working very well at higher memory voltages. P4's seem great also, with the exception of boards with OVP (over voltage protection), as they limit the available voltage to the board. The benefit for those boards will be the filtering circuit.

We are still testing, so hang loose.

Quote:

---

Originally posted by bigtoe

If you think you have a point to make show me where HT voltage is linked to VDD/VDDQ so that I may test it...if you can't please let the tweakers tweak and the ones who are NOT so sure (like you) please avoid this product and run things at stock.

---

So you do want me to prove myself. I'm not the one who wasn't sure. I wasn't the one who asked about it.

If you want to know where it's going into the memorycontroller of most systems, you have to get your hands on the datasheets of the northbridges.

for the memorycontroller of the A64 : grab yourself the datasheets of the A64 on AMD's website and search for the pins of the cpu who are connected to the memory.


cpu interacts with the memory controller and memory controller interacts with the memory. on most of the systems, the memory controller is integrated in the northbridge. on A64 systems, the memory controller is inside the cpu.

all these components have a PLL connected to them. So they can synchronise with eachothers.

when there is a bit in the memory wich represents a logical 1, that's a capacitor wich holds the ddr voltage.

when the memory controller fetches that bit it compares it with it's reference voltage. when it's equal or higher, the memory controller interpretes it as a logical 1. when it's lower or gnd, it's seen as a logical 0.

so when there is memory activity, there are running pulses of ddr voltage between memory and memory controller. these are going very fast. the speed you are running them ... 200 milion times a second.

to actually see those pulses, you need a scope to mesure them. when there is going 2.5V for instance pulsing 200milion times a second, your multimeter probably show you 2.5V all the time. as if it was permanently 2.5V . so just follow the leads from the memory to the chipset. or with an A64, the ones wich are going to the cpu. (or look into the datasheets) and mesure on those pins. With a bit of luck your multimeter shows 2.5V. With less luck it goes tilt. In that case buy yourself a Scope to mesure it. When you are working with digital stuff and things operating at high frequencies, a scope is a must have.


I won't go any further into the fully workings, this is pretty basical explained. But again, it's up to you, to know all of this. Not to me, to show I do.

Before messing with voltages, try to understand what a bit is. And how a pc can make a bit out of a voltage. Don't try to run, when you didn't even learn to walk first.

now, I already spended to much time on this. So you won't see me post again on this subject.

ive studdied the specs some months ago and came to the conclusion that vdimm reaches the integrated memory controller inside the a64 cpus, so raising the vdimm means you raise the voltage that gets bumped through a part of the cpu.


i think you should have a good look on how the vdimm behaves when booting or re-booting the system with the ocz module placed. it was the vdimm fluctuations that killed the a64s, not vdimm itself. i know poeple who are running 3.8v vdimm on a64s for months without any problems.

i think you should add some capacitors so the voltage rises steadily when the system is powered on with the ocz module in place set to increase the vdimm to lets say 3.8v

for the memorycontroller of the A64 : grab yourself the datasheets of the A64 on AMD's website and search for the pins of the cpu who are connected to the memory.

Done that...

cpu interacts with the memory controller and memory controller interacts with the memory. on most of the systems, the memory controller is integrated in the northbridge. on A64 systems, the memory controller is inside the cpu.

Also know that...but it doesn't mean VDDQ/VDD is the applied voltage...its not on the VNF3-250 it seems. Also the memory controller does NOT control Vdimm to the ram...

So to conclude i think we both need to know more about what actually kills A64's..i have read many threads and done heaps of research and the dury is still out of the actual answer.
Many say vdimm, many say 3.3vline, some say VBT, some say Vcore....

any compatability with NF2 boards???

seems like the best 32bit athlon chipset is being left out :(

Quote:

---

Originally posted by saaya
ive studdied the specs some months ago and came to the conclusion that vdimm reaches the integrated memory controller inside the a64 cpus, so raising the vdimm means you raise the voltage that gets bumped through a part of the cpu.


i think you should have a good look on how the vdimm behaves when booting or re-booting the system with the ocz module placed. it was the vdimm fluctuations that killed the a64s, not vdimm itself. i know poeple who are running 3.8v vdimm on a64s for months without any problems.

i think you should add some capacitors so the voltage rises steadily when the system is powered on with the ocz module in place set to increase the vdimm to lets say 3.8v

---

That is all inplace on the final shipping versions..i have said MANY times my unit is early beta if not alpha and as such has been modified to fit on A64 boards.

I can say from my tests today vdimm fed thru the module is extremely stable with zero fluctuation even with out the caps so with them it will be super stable.My module has provision for 10 smoothing capacitors or so on the back...Im sure you will agree once these are fitted voltages should be far smoother than anything the board can deliver.

Quote:

---

Originally posted by slavik
any compatability with NF2 boards???

seems like the best 32bit athlon chipset is being left out :(

---

Its not being left out..we just have not moved onto NF2 yet. Im hoping to get a beta unit to AndyT also so that testing can be spead up a little.

We are beginning extended NF2 testing shortly. So far things look good on that platform.

zabomb4163, OCZ is not recommending that you use this product to over-volt past the point of safety. This means that exceeding your motherboard's specs and memory's recommended voltage will void all warranties. Trying to tell you what is safe for a "mod" and what voltage your memory will blow up at is not going to happen. :)

Quote:

---

Originally posted by bigtoe

---

Quote:

---

Also know that...but it doesn't mean VDDQ/VDD is the applied voltage...its not on the VNF3-250 it seems. Also the memory controller does NOT control Vdimm to the ram...[/B]

---

I didn't say the memory controller controlls the vdimm to the ram.
You don't even seem to understand what I say. I could start crying for the moment out of frustration !

bits flow between memory and memory controller. and between memory controller and cpu.

a bit is a Voltage. Ram stores bits. a logical 0 is GND, thats 0V.
A logical 1 inside the memory is DDR voltage, thats 2.5V stock.

when a memory controller is orderd by the cpu to fetch data at a memory location, the memory controller fetches them from the memory and stores them in the memorycontroller databuffer.

so the memorycontroller by figure of speech says to the memory; hey you scumbag memory send me your bits at that location.
The memory sends those bits to the memory. so the memory sends 0 and 1's to the memory controller. those 1 and 0's are DDRvoltage and 0V.

ofcourse it doesn't go the the nf3 chipset. the nf3 chipset hasn't got a memory controller. the memory controller is inside of the cpu. so those bits travel to the pins on the cpu where it's memory controller is located.
Those pins are on the datasheets.

Also when the memory has send its last bit of that memory location it sends a termination bit to the memory controller. so that the memory controller knows that it doesn't have to wait anymore for other bits.

Quote:

---

So to conclude i think we both need to know more about what actually kills A64's..i have read many threads and done heaps of research and the dury is still out of the actual answer.
Many say vdimm, many say 3.3vline, some say VBT, some say Vcore....

---

what "some" say isn't important. Facts are important.

hey ocz guys lets say you set this gizmo to 3.30v will it stay at 3.3 under a memory intense program or will it droop any?

ns_ripper

You misunderstand me to the point i can scream thru frustration also...


ALL IM BOTHERED ABOUT IS NOT KILLING a64 CPU'S WITH THIS DIMM MODULE...!!!

Now i shouted and I hope you heard. I just read 4 posts of guys running 3.8Vdimm for 3 months on A64 rigs with NO ISSUES..NO dead cpus and NO dead boards.

My I suggest you DON'T buy one of these modules then you WON'T have an issue with it.

Regarding overvolting dimms etc tell Macci et all not to push their rigs hard for the #1 spots on the orb etc..infact tell everyone you like but don't tell me anymore as im bored of it...

You have now entered the realm of th ignore option..Bye Bye :)

Quote:

---

Originally posted by bigtoe
ns_ripper

You misunderstand me to the point i can scream thru frustration also...


ALL IM BOTHERED ABOUT IS NOT KILLING a64 CPU'S WITH THIS DIMM MODULE...!!!

Now i shouted and I hope you heard. I just read 4 posts of guys running 3.8Vdimm for 3 months on A64 rigs with NO ISSUES..NO dead cpus and NO dead boards.

My I suggest you DON'T buy one of these modules then you WON'T have an issue with it.

Regarding overvolting dimms etc tell Macci et all not to push their rigs hard for the #1 spots on the orb etc..infact tell everyone you like but don't tell me anymore as im bored of it...

You have now entered the realm of th ignore option..Bye Bye :)

---

In fact.... isn't 3.2V in the LOW end of Windond's spec's for the BH-5 module?

C

Quote:

---

Originally posted by andyOCZ
We are beginning extended NF2 testing shortly. So far things look good on that platform.

zabomb4163, OCZ is not recommending that you use this product to over-volt past the point of safety. This means that exceeding your motherboard's specs and memory's recommended voltage will void all warranties. Trying to tell you what is safe for a "mod" and what voltage your memory will blow up at is not going to happen. :)

---

He didn't ask for you to advise him on what is safe for his mod, and what voltage his memory will blow up at. He is asking specifically what voltage your memory blew up at in your tests. This is xtreme, and you don't need to "protect" us. If we blow memory using this piece of information, it's not like we're going to sue you or something. Just by posting here you have recognized that we're different from other computer users, and we can handle certain things, like the information that zabomb requested. I see no detriment to you if you give us the information.

The whole idea here is for the device to provide filtered power that is not regulated by the motherboard's voltage regulation circuits. It will not powered by the motherbboard, so your PSU will provide the juice. If you PSU sags, then you may have sag. ;)

Nice ;) Good thing I have a good PSU. I can't wait to get my hands on one. I can see my 3DMark scores rising :P

Right...i think i have found what we needed to know without the bamballings of earlier posts.

Here you will see 2 tables taken from the AMD white papers you can see here.
http://www.amd.com/us-en/assets/cont...docs/31410.pdf

I have tried to show absolote max ratings by AMD.I know nearly all of you work well above this already so this is just to satisfy a few. I will also state you would be best to stay within these values and OCZ do not endorse you using higher voltages than you see here...if you choose to its by your own decission and we have not forced you to do this.

Going by what you see here max Vdimm should be 2.9V.max vcore should be 1.65v.

Very nice indeed. You where talking about the vdimmkontroller should work fine with the -250 chip but how about the old -150 chip? will it work with that as well? I think this is a nice thing from ocz and I hope you will find a way so it could support a64 aswell. Keep us updated.

Quote:

---

Originally posted by bigtoe
ns_ripper

You misunderstand me to the point i can scream thru frustration also...


ALL IM BOTHERED ABOUT IS NOT KILLING a64 CPU'S WITH THIS DIMM MODULE...!!!

Now i shouted and I hope you heard. I just read 4 posts of guys running 3.8Vdimm for 3 months on A64 rigs with NO ISSUES..NO dead cpus and NO dead boards.

My I suggest you DON'T buy one of these modules then you WON'T have an issue with it.

Regarding overvolting dimms etc tell Macci et all not to push their rigs hard for the #1 spots on the orb etc..infact tell everyone you like but don't tell me anymore as im bored of it...

You have now entered the realm of th ignore option..Bye Bye :)

---

You actually work for OCZ? If so, you need help with your PR tactics. This is definitely not the way to try to win people over to your new product.

And if you're research into product compatability is just to search through these boards for people running insane voltages... my bet is you are going to have a few problems when this product is released.

NS may not have come off very well in your eyes... but you should at least get his drift of "make sure you cross your T's and dot your I's" before you release a product to the masses.

:toast:

Quote:

---

Originally posted by andyOCZ
"If you PSU sags, then you may have sag. ;)"

---

Now why does that put scary images in my mind's eye? :D Sounds great tho, about the filtering. I'd rather trust my PCP&C's power regulation than any mobo's. ;)

Quote:

---

Originally posted by Jeff
You actually work for OCZ? If so, you need help with your PR tactics. This is definitely not the way to try to win people over to your new product.

And if you're research into product compatability is just to search through these boards for people running insane voltages... my bet is you are going to have a few problems when this product is released.

NS may not have come off very well in your eyes... but you should at least get his drift of "make sure you cross your T's and dot your I's" before you release a product to the masses.

:toast:

---

I think Tony made it more than clear enough that he hasn't just been looking through these boards. He said he spent what? at least 14 hours looking into this thing...he just posted some stuff from the AMD white paper and other stuff.

OCZ didn't have to pre-post this here to show everyone what cool things they are bringing out so just feel happy they showed you and stop smashing them. I don't see any other companies here showing us the cool things they are making for us....let alone before they go out to sale. The product isn't out yet and they have time to test everything that needs testing I would assume.

So lets all chill and be happy and wait to see how this performs when all the good people here get it :toast:

Steve

yepp, calm down, bigtoe is just testing it, hes not the one who invented it or builds the final units you will be able to buy, ergo, he doesnt have to know how exactly it works on every specific chipset out there, thats not his job.

:)

Maybe there's a reason companies don't post on public forums.

Not being a wise-ass but if you are going to do this sort of thing... either make it clear from the beginning that you will NOT be answering any questions about the product... OR... have someone fielding questions that is qualified or at least knows when to say "I'm not familiar with that aspect of the product... forgive me but I'm going to have to have one of my co-workers look into that for me."

Yep... it's nice having a company share but at least have some ground rules and don't have someone in an official capacity say "Yep, you're on my ignore list now buddy because I'm frustrated with you!"

See my point? (kinda-sorta maybe?)

At any rate... no matter what my thoughts are on this product... I will be buying one since it's another toy to play with. Not going to be using it on my A64 to start off, but my NF7-S and 8RDA+ should be fun. :) (may even try my KT400 and KT600 boards too ;))

hmmm yeah jeff i totally get your point... (i think lol :D)

but on the other hand i think bigtoe just wanted to share this info with us and wanted to help by answering those questions...

and i think the main reason why companies dont post on public forums is because they try to hide from critics. asus for example would never admit that some of their products has only been tested in a few configurations, instead they dont comment on it at all.

bigtoe instead amdited that they have only tested this device on a few systems yet, but thats no problem as its still in development, and i dont think they will claim to have tested it on any system that they havent tested it on...

i dont really understand the problem some of you seem to have here. instead of saying HOORAY we now have a way to get a higher vdimm without modding the board on those boards:

P4P800E
P4C800E
IC7
IC7max3
DFI 875B

people complain about what could happen on a64 boards and other boards that are not listed yet...

ocz is still a rather small company, and you cant really expect them to release a product that has been tested on all available systems. even the largest manufacturers release really crappy products from time to time that havent been tested at all, the diference is that they hide it...

if it works for you : HOORAY!
if it doesnt work for you : what do you lose?

you asked him to test it on a64 platforms and he goes and tests it on a64 platforms... you say test it on nforce2 platforms and he says ok, i will... tell me only one other company that does this!

if they find out it doesnt work on some platforms they will put a note about it on their page and on the product description and probably try to find a workaround in the final version or for version 2.0 ...
i think some of you are just in a bad mood and looking at it in a negative way instead of a positive way :)

so ... when is it going to be in retail?

Alright boys, be nice! LOL

bigtoe was being taken down the road of contrary Sunday fun. He is only trying to answer questions. When the questions become a turkey shoot with him as the target, you see what happens. We come in here to try and let our friends have a heads up on products. IMHO it is a good thing.

And yes, most "official" support folks won't brave the Xtremes of Xtreme. bigtoe and I do work for OCZ and try and keep a cool head. Why is it that we are some of the few that will take you guys on? Last time I checked our work hours end on Friday and start up again on Monday. So Sunday is "off time" ;)

Retail will be a few weeks off at least from what I have been told.

Have a good one!!

Quote:

---

Originally posted by andyOCZ
We come in here to try and let our friends have a heads up on products

---

and your very welcome! :toast: (at least for most people here hehe :D)

must be hard to switch between the enthusiast and business world, i dont envy you guys for nothing! :D

bigtoe, just use the "no comment" reply more often next time :D hehe

Quote:

---

When the questions become a turkey shoot with him as the target, you see what happens.

---

Well, when he comes in here to answer questions, and he ignores or refuses to answer some of them, this is what happens.

EDIT: by the way, I'm referring to my request for patent information (ignored), and zabomb's question about voltages and blowing chips (refused). I don't like the way he refused it, either, telling him he didn't want to know.

in the past ocz has been more open with customers :(

yeah im also curious about how many volts fried that eb memory :D

but i dont think it matters, i think theres a large gap between the volts you dont gain any mhz from increasing vdimm and the volts you need to reallyfry the chips.

and about the patent, what do you mean?

Quote:

---

Originally posted by saaya
yeah im also curious about how many volts fried that eb memory :D

but i dont think it matters, i think theres a large gap between the volts you dont gain any mhz from increasing vdimm and the volts you need to reallyfry the chips.

and about the patent, what do you mean?

---

First bigtoe said it was patent applied for, then he said it was patented. Zabomb asked whether there was a patent or if it was only applied for, and I asked for a link to the patent so I could read what it said, but we were ignored.

Quote:

---

Originally posted by squeakygeek
First bigtoe said it was patent applied for, then he said it was patented. Zabomb asked whether there was a patent or if it was only applied for, and I asked for a link to the patent so I could read what it said, but we were ignored.

---

You should give him ample time to respond before making such a comment sir. :idea:

Not likely he is ignoring you.

He started this thread and I am sure he plans to answer peoples questions or he wouldnt have made such a thread knowing that it would stir up peoples curiousity.

Quote:

---

Originally posted by Hell-Fire
You should give him ample time to respond before making such a comment sir. :idea:

Not likely he is ignoring you.

He started this thread and I am sure he plans to answer peoples questions or he wouldnt have made such a thread knowing that it would stir up peoples curiousity.

---

Well I asked 3 pages ago. He has made many posts since then with no indication that he planned to give me the information. If he just hasn't gotten around to it yet, I apologize, but if he would have acknowledged my question, and told me he needed more time, we wouldn't be having this misunderstanding. If ocz has a patent on this, it should be very easy to at least give a patent number.

maybe you should try giving them a pm.

If the device has a provisional patent then there will be no info gained from the number. Big Toe is here to inform us on a new device he is working on. Is it proper to ask for proprietary information. I don't think it is. Any associated patent is his to disclose or not.

Why do you want the Patent #?

Quote:

---

Originally posted by chilly1
If the device has a provisional patent then there will be no info gained from the number. Big Toe is here to inform us on a new device he is working on. Is it proper to ask for proprietary information. I don't think it is. Any associated patent is his to disclose or not.

---

I don't know much about patents, but from a quick search, I get the impression that there isn't really such thing as a "provisional patent". A "provisional application" for a patent can get you "patent pending" status for 12 months, but it's not a patent.

I was also under the impression that patents are public information: http://www.uspto.gov/patft/index.html It wouldn't make much sense for patents to be secret! I didn't ask him to explain the inner working of his device to me, I just want to know whether there is in fact a patent, and if so, what it covers. And since he was the one who brought up the patent, I think it is proper to ask what I asked, and improper for him to ignore me, if he did in fact mean to ignore me.

Quote:

---

Originally posted by spaceman
Why do you want the Patent #?

---

... because I want to see the patent, why else?

The real question is why didn't he give me one?

some info on patents on provisional applications for patents: http://www.uspto.gov/web/offices/pac/provapp.htm

Looky here guys. Let's get real. bigtoe and I work in a support/development role, we don't work in the front office. I know next to zero about US Patent law. bigtoe lives in England and knows less. If you want a copy of all patents applied for or granted to OCZ, contact OCZ through the website. If the lawyers feel you have a need for them they might give you the list. If not this is public information. Contact the US Patent Office. Thank you.

bigtoe did not start this thread to dodge your questions. He took time out his day off to answer some questions. If he missed some, I'm sure it was not intentional. Read the posts in between your questions. He was being badgered and was trying to be reasonable, at least for a while. If you want to discourage these threads this is a good way to do it.

I will attempt to answer your questions in another post.

ocz rocks...........

Question: Is this product patented?
I was told the patent request was filed. I know nothing more about it. It is unlikely bigtoe or I will find out more. It ain't our job. :)

How many volts did it take to fry the EB RAM?
5 volts. It was accidentally hooked straight to the 5 volt rail. The bread board prototype was being "tested". (really)

How many volts will it take to fry every memory module that might be subjected to the unit?
We don't know. We will not be blowing up a module of every RAM chip made for the last 3 years and a board to go with it to find out. Use some common sense, or you may find out. Winbond says 3.6v is the maximum for some of it's popular IC's. I have been told that, but have not checked for myself. You might look at the white paper for your ram IC's from the IC manufacturer to find the max voltage.

Is the design complete for the device?
Yes

Has every board on every platform been tested at this time?
No, but we are working on it.

Will this product void my warranty if I push over the limits as stated by the memory and board manufacturers recommendations? Yes, most certainly.

If I were to do a volt mod on my board instead of using this product, like many of us already do, will that void my warranty? YES

Why is bigtoe not posting the answers to my questions as soon as I post?
To start with bigtoe "works" Monday to Friday. This is Sunday. All posting is done in our spare time anyway. Our job is not just posting. Hey, the guy lives in England. It's 9:35pm here and 5:30 in the morning there.

Is bigtoe a patent attourney? NO

I do think that ns_ripper's was pretty inappropriate

why he felt the need to lecture OCZ is beyond me

most members here can be considered electronics hobbyists, nothing is new, and he's suggesting to OCZ that they refer to the AMD white papers, what a joke

well, he may be underestimating their efforts, but these modules do in fact seem to have an advantage over a traditional vdimm mod, and it makes them very interesting, but I think I do share his suspicions on how these will work in the open market.

Quote:

---

Originally posted by squeakygeek
... because I want to see the patent, why else?

The real question is why didn't he give me one?

---

No, the REAL question here is, where does a guy with 27 posts get off bustin on guys, who are tryin to do us a favor? Get over yourself, and show a little respect. :flame:

That's why the units are not available retail yet. We have been testing for a while now, and are testing every available board for compatabilty.

Quote:

---

Originally posted by spaceman
No, the REAL question here is, where does a guy with 27 posts get off bustin on guys, who are tryin to do us a favor? Get over yourself, and show a little respect. :flame:

---

I wasn't busting on anyone, and I don't think I've disrespected anyone... and I don't think post count has anything to do with this... you callin me a n00b or something :p

Quote:

---

How many volts did it take to fry the EB RAM?
5 volts. It was accidentally hooked staright to the 5 volt rail. The bread board prototype was being "tested". (really)

---

Thanks.

You are welcome.

well if you guys are taking any recommendations on which boards to test next, maybe the msi k8n neo platinum :D

So far all socket 754 boards are pretty much the same, meaning it works fine.

Any testing on MSI 865Neo 2?

I am glad to see this calming down. As with all questions fielded on these boards no one is under any obligation to answer. Andy thanks for the explanation, And also thanks for letting us in on your current developments. I hope that this does not discourge you from future revlations.

Please keep this on topic...

Quote:

---

Originally posted by andyOCZ
So far all socket 754 boards are pretty much the same, meaning it works fine.

---

:D :banana: :banana:
thats good to hear

Quote:

---

Originally posted by andyOCZ
How many volts did it take to fry the EB RAM?
5 volts. It was accidentally hooked straight to the 5 volt rail.

---

thx! :toast:

lol you dont happen to have any pics of that stick of memory do you? :D and the board survived it? amazing!

so using the device to raise vdimm to 3.2v is not voiding the warranty if the board supports 3.2v in bios as well right? but it will 8eventually) give a cleaner 3.2v vdimm than the board can produce?

and boards with an vdimm OVP will only be able to get vdimm up until the OVP kicks in? so it doesnt matter for the OVP wether the vdimm comes from the board or the ocz device.

So far those things seem true. The motherboard warranty is always a question. I can't answer for them. ;)

The filtering circuits and a more stable power source should help with keeping the system stable.

OCZ since the PC3500 Limited Edition days has been my favorite computer component company. Quality parts that cator to us, the enthusiasts. :)

I love you guys, keep up the great work and never let a few forum punks let you down. Anyone who's anyone is happy about this. ;)

Thanks, it's not a problem. We all have questions. I sometimes wonder where they are leading, but humans is humans. ;)

squeakygeek

Its not my place to show you the patent forms, if Ryan wants to do this then that is upto him.It was applied for a good while ago and should be going thru real soon.

Also im on the inventors list for part of the technology behind the idea's used on the dimm, it also covers other things also.

ns_ripper was dodging round my question just as much as i was dodging round his..the question was "does excessive vdimm kill A64's ?"...at no point did i see him say "yes"

Lastly

OCZ regards a few places on the net home, Xtreme is Ryans favourite forum and as such he likes to let you guys in on a few sneaky tidbits before anyone else.That is why we leaked early info on the dimm here.

OCZ are here for the overclocker, we are trying to give the overclocker just what they want.We had completely dropped BH5 and BH6 but a flood of emails etc had us relaunching the modules.We had requests for a good PSU with adjustable rails and we came thru on that one also.
We search for the memory IC that gets you that little bit more even if we can only see 3 months run as it helps the enthusiast get that little more from their rig...Corsair have played on this saying we only supply special edition modules but they forget to look at the rest of our product line and see the same modules have been available for a year or more :confused:

So...we could become a boring ram manufacturer supplying mudane modules or we can stay at the cutting edge trying to deliver what you guys want for your rigs....

Quote:

---

Originally posted by bigtoe

http://www.amd.com/us-en/assets/cont...docs/31410.pdf

---

You know I find this very interesting, VDIMM is 2.9v max but VBT/VTT is allowed to go up to 1.65v. VBT is normally half VDIMM so there is some extra leeway here it seems

I used to run my old Abit KV8 Max3 at 3.2v no problem and at this speed VBT would still be in spec, it would only be out of spec at above 3.3v.

I wonder if VBT is the real killer of AMD64's, if it is then that would indicate 3.4v and above is the danger area. Of course not all boards track VBT with VDIMM and for them it would not be an issue ( but then it is unstable anyway ).

Funnily enough my old KV8 MAX3 did track VBT as far as I can remember, my new KV8 Pro does not, and my new KV8 Max I am not sure as it locks as soon as you select above 2.8v. I think the motherboard manufacturers and AMD have been examining all those dud AMD64's from XS :) and put the brakes on.

I get the feeling the Shuttle allowed VBT tracking and hence the problems with over 1.65v VBT, also probably why it was so good at overclocking.

I also think Gigabyte K8N did not kill AMD64's but their VBT did not track VDIMM .. Macci changed his VBT on this board but only lost a cpu when his VBT doubled for some reason.

In regards to this OCZ widget with AMD64 if my rather haphazard guess is right then if VBT does not track VDIMM then no trouble, but then not much use, if it does track then I think I'd be happy with this device up to 3.3v, which for me is enough anyhow.

Thanks for the info on this, hopefully this rather active discussion has helped people prepare for potential pitfalls, those AMD64's are fairly expensive after all.

Any chance of testing one on the MSI NEO, or alternatively does anyone know any AMD64 motherboard where VBT does track ?

Regards

Andy

Tony, what u think of using this on the new socket 939 mobos?

... im thinking of asus A8v ;)

So am I right in saying that this will work properly on some boards but other boards will still require another mod?? Also and it is somewhat off topic but are the OCZ Power Supplies available yet? I havnt been able to find any yet at my main retailers here in Canada that usually stock all my favourite OCZ products :)

Quote:

---

Originally posted by bigtoe
squeakygeek

Its not my place to show you the patent forms, if Ryan wants to do this then that is upto him.It was applied for a good while ago and should be going thru real soon.

Also im on the inventors list for part of the technology behind the idea's used on the dimm, it also covers other things also.

ns_ripper was dodging round my question just as much as i was dodging round his..the question was "does excessive vdimm kill A64's ?"...at no point did i see him say "yes"

Lastly

OCZ regards a few places on the net home, Xtreme is Ryans favourite forum and as such he likes to let you guys in on a few sneaky tidbits before anyone else.That is why we leaked early info on the dimm here.

OCZ are here for the overclocker, we are trying to give the overclocker just what they want.We had completely dropped BH5 and BH6 but a flood of emails etc had us relaunching the modules.We had requests for a good PSU with adjustable rails and we came thru on that one also.
We search for the memory IC that gets you that little bit more even if we can only see 3 months run as it helps the enthusiast get that little more from their rig...Corsair have played on this saying we only supply special edition modules but they forget to look at the rest of our product line and see the same modules have been available for a year or more :confused:

So...we could become a boring ram manufacturer supplying mudane modules or we can stay at the cutting edge trying to deliver what you guys want for your rigs....

---

:thumbsup:

hmmm but if amd says 1.65v vbt/vtt is the max it doesnt mean 1.7v will kill an a64. they usually leave a large headroom to be sure nothing goes wrong. so i guess even 1.9v and maybe even 2v vbt/vtt is not going to kill an a64.

just like every manufacturer says that 2.9v or even 2.85v is the max vdimm for their memory while no memory has died from even 3.5v yet.

VTT as per the data sheat is termination voltage for VDD (CPU vcore) on each side of the cpu and is NOT linked to VTT on the dimms.Vref(ram) does go to the cpu as does it seems VDD(ram).

Macci increased VTT...VBT is not a term i tend to use. VDD/VDDQ Vref and VTT are all ram related with VTT hitting the board and not the dimm.The ONLY voltages changed on the dimm are VDD/VDDQ and Vref.VTT has to track Vref for the dimms to perform correctly.

So looking at the data sheet VTT can hit 1.65V which must mean it should be equal to VDD Cpu vcore...so in theory its a little similar to Vref on a dimm being equal to VTT termaination voltage except this time is vcore=VTT.

Looking at the VNf3 250 it does acually seem Chipset voltage is infact HT voltage and with 1.9V available we already have a huge overvolt capability.

As I uncover more i will post, I am asking questions that should answer all we need to know about A64 etc as it seems no one actually has the definitive answer to all the questions we have been seeing.

Quote:

---

Originally posted by bigtoe
squeakygeek

So...we could become a boring ram manufacturer supplying mudane modules or we can stay at the cutting edge trying to deliver what you guys want for your rigs....

---

Likewise...:thumbsup:

Geez guys why is anyone busting on Bigtoe. If you go back and read NS's first post he was over the line. I thought Bigtoe showed great restraint. Anyway thanks to OCZ for another great product. You guys rock

I wouldnt say he was over the line, he just tripped and fumbled the ball.. ;)

Guys............guys..............Take it easy...........This company(OCZ) manufactured something for us overclockers and they spend a lot of their time so as to design and to put it together.............Who else did that???........I see that this is a great help for us "crazy" overclockers.........I manufacture myself (Radio and TV transmiters) and I know how DIFFICULT is to design something from the begining and to test it - redesign and retest it all over again and so on , so as to be ready for the public.....
It's not easy.........and it's NOT a ram module which is similar with the others........It's a BRAND NEW DESIGN........

ALSO they didn't design it and test it so as to give it to as that quick.......They retesting it again and again so as to be sure IF it's OK and ready or not for the public..... ;)

Try to be pattient and see what they'll do........

I think that OCZ makes a step ahead because they "think" about us........WHY no other company did something else except of starting a "war" about "how fast is my ram module - I beat you all" , "I can do 275MHz with only 2.8Vdimm" (didn't mention the timmings for the 275MHz though - Cas 7 - 16 - 16 - 32) :D

So in my opinion we have to wait and see what will happen...... ;) :toast:

Looks like a great product :slobber:

When can we expect the product in Europe?

SHI(F)T...........I have read the whole thread carefully(about 1.5 hours now -shi(f)t)........ns_ripper
...........You now(or have read) a lot theory BUT you are WRONG - as for me.........I'm sorry to say that BUT you are wrong........shi(f)t.........The bits passing through lines and internal Mos-Fets are NOT passing through capassitors(charging-uncharging) as you give your examples.......They pass through the Mos-fets which are "switchers" ON and OFF......So there is a perfect FLAT LINE bit at the end.........In worst cases that flat line has some "spikes" on it from "external" or "internal" reasons............Shi(f)t........and IF you want to show you some testing equipments of mine ,I'd be delightful (HP(Hewlett-Packard)-Tektronix-Marconi-Advantest-Philips-Hameg-e.t.c.)....I can't read those things any more .......I'm sorry I can't handle myshelf when I read things like that....... :eek:

..........and PLEASE guys when you read a manual,you'd better read it from the begeining till the end........ ;)

........at stress conditions(page 95).......Thank you........

Bigtoe send this man (hipro5) one of these modules to test :D

even a preproduction sample...

i am sure we will see amazing things ;)

Ãéùñãï êåñÃ:banana:áò ìðõñåò áÃ:banana: óïõ'ñ8åé...

Quote:

---

Originally posted by brwmogazos
Bigtoe send this man (hipro5) one of these modules to test :D

even a preproduction sample...

i am sure we will see amazing things ;)

Ãéùñãï êåñÃ:banana:áò ìðõñåò áÃ:banana: óïõ'ñ8åé...

---

Unfortunatelly bigtoe keeps "holding" hard feelings about me , so.......

Well.

VDDIO covers the Ram VDD which jedec quotes as 2.6v so my absolute max of 2.9V holds true.

VTT cpu is allowed to hit Vref(ram)+0.3V so this means to me 1.5V VTT+0.3V max for 1.8V VTT CPU max.

VLDT is allowed to be overvolted by 0.3V max which holds true on the VNF3 250 with 1.9V chipset voltage = VLDT.

So nothing to me altered from what i posted earlier on except for VTT CPU which I reported at 1.65V max but does seem to be 1.8V max.



Thanks for the help Hipro5 :)

Quote:

---

Originally posted by hipro5
SHI(F)T...........I have read the whole thread carefully(about 1.5 hours now -shi(f)t)........ns_ripper
...........You now(or have read) a lot theory BUT you are WRONG - as for me.........I'm sorry to say that BUT you are wrong........shi(f)t.........The bits passing through lines and internal Mos-Fets are NOT passing through capassitors(charging-uncharging) as you give your examples.......They pass through the Mos-fets which are "switchers" ON and OFF......So there is a perfect FLAT LINE bit at the end.........In worst cases that flat line has some "spikes" on it from "external" or "internal" reasons............Shi(f)t........and IF you want to show you some testing equipments of mine ,I'd be delightful (HP(Hewlett-Packard)-Tektronix-Marconi-Advantest-Philips-Hameg-e.t.c.)....I can't read those things any more .......I'm sorry I can't handle myshelf when I read things like that....... :eek:

---

no, in all respect, the ram chips themselfs are slightly capacitive. ns_ripper is correct about those peaks. i think corsair had a demonstration once it should still be on their website.

Quote:

---

Originally posted by enzoR
no, in all respect, the ram chips themselfs are slightly capacitive. ns_ripper is correct about those peaks. i think corsair had a demonstration once it should still be on their website.

---

Everything in electronics have their capacitence.......Everything........but we are talking about 1 and 0 ....on and off....Not 1.12 and 0.04 or 0.97 and -0.1.....1 and 0...... :)

Quote:

---

Originally posted by hipro5
Unfortunatelly bigtoe keeps "holding" hard feelings about me , so.......

---

I have no hard feeling m8...so please stop thinking i do.

As a moderator of a forum I asked you to do something that members had requested..nothing more.

Quote:

---

Originally posted by bigtoe
I have no hard feeling m8...so please stop thinking i do.

As a moderator of a forum I asked you to do something that members had requested..nothing more.

---

I'm talking about "Hawk's fiasco"......Nothing more , nothing less.....

Quote:

---

Originally posted by hipro5
I'm talking about "Hawk's fiasco"......Nothing more , nothing less.....

---

That was a bios m8...seen the same with the MSI letting you set 2-2-2-5 at 270fsb when in reality your running 3-4-4-8.

I also have the bios Hawk was using...hole in the fsb range from 165 to 199 although Hawks board was a very early sample that ran above 166 where the others crashed.It however was NOT running the speed he was setting in bios...this again has happened on the MSI etc.

So..lets drop it...i know you were partially right but Hawk was not a scammer, the bios and board was the scam not the user.Remember he had a preproduction board which may act differently with different bios files.

OCZ does it again! I'll be gulping one down for you gents tonight :D

I can finally breathe new life into my aging 845PE system that my BH5 couldn't help due to lack of voltage.

Keep up the good work!

Quote:

---

Originally posted by bigtoe
That was a bios m8...seen the same with the MSI letting you set 2-2-2-5 at 270fsb when in reality your running 3-4-4-8.

I also have the bios Hawk was using...hole in the fsb range from 165 to 199 although Hawks board was a very early sample that ran above 166 where the others crashed.It however was NOT running the speed he was setting in bios...this again has happened on the MSI etc.

So..lets drop it...i know you were partially right but Hawk was not a scammer, the bios and board was the scam not the user.Remember he had a preproduction board which may act differently with different bios files.

---

OK.......Let's give an end here..........Don't want to remember.... :)

Quote:

---

Originally posted by enzoR
no, in all respect, the ram chips themselfs are slightly capacitive. ns_ripper is correct about those peaks. i think corsair had a demonstration once it should still be on their website.

---

ns_ripper was talking about the SMD capacitors on the traces which are there to eliminate/smooth ripple and help remove trace interferrance.

Hipro is correct, mosfets (look up p and n mosfets on google ;) ) are what are producing the signal, and they do it very well, else your CPU would not function.

QUOTE]Originally posted by STEvil
ns_ripper was talking about the SMD capacitors on the traces which are there to eliminate/smooth ripple and help remove trace interferrance.

Hipro is correct, mosfets (look up p and n mosfets on google ;) ) are what are producing the signal, and they do it very well, else your CPU would not function. [/QUOTE]

mustave mis-understood his post then... happens with me sometimes, but iirc, the "peaks" that ns_ripper posted are correct. the ram chip itself is resistive and slightly capacitive so it takes a while for the signal to switch from on and off. its not straight ON right away, theirs a curve. when it goes to off, because the chips is capacitive, it does not go OFF right away, hence another curve. increasing voltage overcomes this.

or am i totally wrong? :D

ps, hipro5 m8, hows the mod going ;) :p:

cheers enzor
:toast:

wrong, yet correct.

Let me explain it to you this way: If you have insufficient current/voltage you get bad data due the "curving" of the signal.

A good example is fans. They require a startup voltage to even start spinning.. after that they dont take quite as much. Due to trace capacitance and other items this is pretty well evened out by the time it hits the chips and the chips, due to being based on mosfets built on a .09 to .15 micron build process rather than the normal 1-5mm accross you are used to seeing can handle the voltage, capacitance, etc with ease.. at default.

When you overclock or change nominal values its always a crapshoot.. :D

Any word on the A64's, whats the max vdimm some of you A64 users are running? Maybe I'll get both at the same time.

I dont think vdimm has to do with them dieing..

A64 death thread in the AMD section has a few people saying 3.3 - 3.6 for 24/7 so it is beginning to lean more towards VBT as the cause of death.

Quote:

---

Originally posted by enzoR
ps, hipro5 m8, hows the mod going ;) :p:

cheers enzor
:toast: [/B]

---

:D

It's ready..... :D :toast:
BUT we have to put heatsinks on the Vmem Mos-Fets from now on.....They are getting too hot...like the AGP ones...We can now feed our rams up to 4.8Vmem(Vdimm).....I'll get a digicam and I'll post..That was TOO easy M8.....One leg lifted up and ....there you go.....5Volts.... ;) :) :toast:

ohhohoho!!! :banana: :banana:

so the +3.3v standby voltage was not a problem? and no need for diodes :banana: :D

Quote:

---

Originally posted by STEvil
I dont think vdimm has to do with them dieing..

A64 death thread in the AMD section has a few people saying 3.3 - 3.6 for 24/7 so it is beginning to lean more towards VBT as the cause of death.

---

Well......Look at what I've said a long time ago about AMD dieing..... ;)

That's what was said by me at 08-19-2002...This S.N.D.S. will very SOON be S.A.D.S.(Sudden AMD Death Syndrome) as well and YOU WILL SEE IT....

Quote:

---

Originally posted by enzoR
ohhohoho!!! :banana: :banana:

so the +3.3v standby voltage was not a problem? and no need for diodes :banana: :D

---

C O R E C T :D :banana: