im thinking of picking up a 3500+ and the A8V myself. I'll stay tuned on this post and see what else develops.
im thinking of picking up a 3500+ and the A8V myself. I'll stay tuned on this post and see what else develops.
I'm running some lvl 2 atm. Don't have any eb.Quote:
Originally posted by WesM63
Ahh I see.. Well good luck with it :D
(What rams are you using? Mushkin lvl2? or OCZ EB?)
Cool, Didn't know for sure.
I'm trying to get a feeling for what I should expect out of this 3700EB i'm ordering monday.
I've just got a 3500 and an A8V!
Best I can get 3d to run is at 220.
Have benched super pi at 11x241 and got 32secs.
I've even loaded windows at 10x250 and my hard drive survived!
I hope a bios update will sort this as Its a good board.
Feels nearly as smooth as a hyperthreaded P4.
legion u pick those up from cpu city?Quote:
Originally posted by legion
I've just got a 3500 and an A8V!
Best I can get 3d to run is at 220.
Have benched super pi at 11x241 and got 32secs.
I've even loaded windows at 10x250 and my hard drive survived!
I hope a bios update will sort this as Its a good board.
Feels nearly as smooth as a hyperthreaded P4.
you have a nvidia card legion? I tried 240x10 and a divider and 3dmark crashed right on video start up. R9800xt couldnt take that much agp bus i guess since i can bench at 220x11
Got an A8V but still waiting on some other parts before building. Reading through the manual, the IDE port on the Promise controller is a little confusing. Maybe one of you guys can answer this for me:
There is a Promise BIOS option with two selections...RAID and IDE. Some parts of the manual make it sound like the option is associated with the IDE port only, others it sounds like it's associated with the controller. What I want to know is this... is it possible to have two SATA drives operating in RAID 0 on the Promise SATA ports, and at the same time have a single independent IDE drive connected to the Promise IDE port?
I'm going to have 2 x 74GB Raptors on the VIA controller, 2 x 74GB Raptors on the Promise controller, and then two optical drives, one on each of the IDE ports. Then I'm going to have a 200GB IDE drive, and don't know if it can be on the Promise IDE port, or if two devices are going to have to share a cable on one of the other IDE ports.
Thanks.
Here's a compare url for 2001se at 235 x 12.5. Not sure why it says fsb = 200. This is on a 9800np stock, sliders at default. Its a 1k better than my 3400+ at 2871 261 1/1 using the same cards and settings.
http://service.futuremark.com/compare?2k1=7901935
I changed the devider and am running at 250x11 @ 1.8v
scored a 61,300+ in Aquamark :)
Outstanding, I was wondering if the divider was causing some problems :toast:Quote:
Originally posted by Rukee
I changed the devider and am running at 250x11 @ 1.8v
scored a 61,300+ in Aquamark :)
Join the forum at Bleedinedge.com and you have access to my bios bin..i just posted 1005.020 if you don't already have it..
This bios fixes a LOT of issues inc the fx53 issue.
http://www.bleedinedge.com/forum/sho...05&postcount=3
got it. thanks., What is the FX53 issue exactly? Maybe that's why I couldn't get the bios to save anything?
He means 1005.020 <<makes no sense now cos you edited your post :(
Had nout to do, so I pimped it from that site.
No idea what it's like etc etc, but here you go :)
Finally got my system up and running. I'm not sure how I did it, but I now have the AGP/PCI lock option in the BIOS. After installing the board, I flashed to the 1005 version that everyone else was using. I then re-flashed using Bigtoe's version to see if it was any different. Anyway, the AGP/PCI Frequency option is now there. :)
Still playing around with dividers. Been able to get up to 250FSB so far.
http://img13.photobucket.com/albums/...PUZ_250X10.jpg
I heard the early boards will not support pci lock but new production models will. Aparantly there are 2 resistors that need adding to early boards although on the pics i have I can't see where.
The 1005.020 is the bios that unleashes this board, there WAS an issue with FX53 this bios sorted..in doing so it helped the other cpu's also.
Next bios aparantly may support more vdimm..;)
Thank god!Quote:
Originally posted by bigtoe
Next bios aparantly may support more vdimm..;)
What was the issue with the FX-53?
What divider are you talking about? FSB:Mem ratio? Wouldn't there have to be a working lock if people are running 250MHz FSB?
It would be nice if we could post the older serials range to be carefull for. Asus should have never sent those to market.Quote:
Originally posted by bigtoe
I heard the early boards will not support pci lock but new production models will. Aparantly there are 2 resistors that need adding to early boards although on the pics i have I can't see where.
I`m up to 256fsb x 11
Might try Bigtoe`s bios soon. :)
http://www.justwebit.com/sites1/42124/256mhz.jpg
Ditto. I ordered my board the day it became available at Newegg. If it's not going to support the lock, I'm going to RMA it. Just need to know how to tell whether it will or not.Quote:
Originally posted by xgman
It would be nice if we could post the older serials range to be carefull for. Asus should have never sent those to market.
Not if they are running dividers, yes fsb/mem ratios, but it looks like the new bios puts the pci/agp locks back in the bios.Quote:
Originally posted by stevenmh
What divider are you talking about? FSB:Mem ratio? Wouldn't there have to be a working lock if people are running 250MHz FSB?
Hard to tell....I will try and find out more. I do have pics but nothing shows...Quote:
Originally posted by xgman
It would be nice if we could post the older serials range to be carefull for. Asus should have never sent those to market.
i have to assume that .020 should be a .bin. for some reason there was no extension on the file
just rename it .rom
I am trying to uderstand this. If Sierra's board after flashing with the 1005.020 now has the locks option in bios, but he still can't really run over 220 1-1 but can run 250 4-3, then doesn't that say the locks, although in bios options, don't work?
I thought the AGP/PCI clocks were based on the FSB clock...so even if you run the memory at a divider, wouldn't 250MHz FSB still take the AGP/PCI speed way out of spec without either a lock or AGP/PCI divider?
And even with the new BIOS adding the AGP/PCI lock back in, what about early production boards without these two resistors? Has this been confirmed, or just a rumor?
Just visited the Asus website and still only see the 1003 BIOS...is the 1005 a beta? Is the one being linked to a 'stock' version or a modded bios?
I'm putting together A8V/FX-53 systems for myself and my brother...I'd definitely like to avoid whatever problems are referenced here....can't find anyone having a specific issue with the FX-53 on the A8V, though.
If someone using the 020 bios can prove that the lock is working and post their serial #, then we can assume that any serials at least after this one or ones will be fixed.
It's a beta bios. Asus always runs these betas in a series and then after awile, releases the last one as final.Quote:
Originally posted by stevenmh
I thought the AGP/PCI clocks were based on the FSB clock...so even if you run the memory at a divider, wouldn't 250MHz FSB still take the AGP/PCI speed way out of spec without either a lock or AGP/PCI divider?
And even with the new BIOS adding the AGP/PCI lock back in, what about early production boards without these two resistors? Has this been confirmed, or just a rumor?
Just visited the Asus website and still only see the 1003 BIOS...is the 1005 a beta? Is the one being linked to a 'stock' version or a modded bios?
I'm putting together A8V/FX-53 systems for myself and my brother...I'd definitely like to avoid whatever problems are referenced here....can't find any specifics on any forums on anyone having problems, though.
And as far as the divider, yes at some point.
we have locks people, in bios anyway. testing soon but first i need food.
no serial no. sticker on the board that i can see. but the serial on the box is 45MGOJ2946
The serial sticker for mine is on the parallel port connector.
45MGOJ2986...so if the lock works on yours, I should be OK.
Just sent in a couple questions to Asus tech support...will post the response here when/if I receive one.
************************************
Dear Asus Representative,
I recently purchased the A8V Deluxe board for use with an FX-53 processor. I haven't built the new system yet because I'm still waiting on a few parts, but I've read the manual thoroughly and am participating in several message board threads specific to the A8V. I have a couple of questions:
(1) I've read that the PCI/AGP lock doesn't work with the 1003 BIOS, but that the 1005 beta BIOS fixes this problem. However, I've also read that early production boards won't support this feature even with a fixed BIOS, due to requiring a couple of extra resistors that are being provided on newer boards. Can you tell me if this is true or not? If so, can you provide a serial number range to differientiate between the boards with and without the hardware to support a working lock? I purchased this board with the understanding that it would have this feature, and the manual clearly states that the feature is present, so if my board does not have it working I will need to RMA it.
(2) The manual is a little confusing on the use of the Promise IDE port. Is it possible to use the two Promise SATA ports in a RAID 0 configuration, and at the same time use the Promise IDE port for an independent IDE drive? The BIOS has an option for 'RAID' or 'IDE' under the Promise menu item, but it's not clear whether it affects just the IDE port or all ports on the Promise controller.
This is my first Asus board and it looks like it's an excellent choice. I can't wait to get it up and running. Thanks in advance for your help with the above two issues.
It's also on the side of the box. But again, we need someone's board that can prove the locks work.Quote:
Originally posted by stevenmh
The serial sticker for mine is on the parallel port connector.
im confident the locks are working. my 9800xt HATES any agp bus higher than about 74~75. i couldnt run 3dmark01se at 240x10 5:3(5:6 i think this asus system is stupid) with a dlt of 4. couldnt get past car low and wouldnt even post video on occasion.
3dmark01se has finished but the score sucked. lower than 215 1:1. aparently you need to get quite a increase in fsb to make up for the divider hit.
I'm still a little confused on the divider issue. Why are you running a divider...because your memory craps out before your board, or because the lock doesn't work with 1:1 even if your memory can handle it? In other words, if I have memory that will do 250, will the lock work if I set my FSB at 250 with 1:1 ratio?
my bh-5 wouldnt do over 220 before. so the only way to really test the locks would be to run a divider since 220/3 is only
73mhz agp, and ive never seen a ati card that would crap out as low as mine did. And my card would crap out with the divider before at that speed.
I dont see how the divider would make a diffrence since the last i knew your memory speed was a product of your fsb and not the other way around:confused:
I'll try 225x11 1:1 just in case my card likes agp bus less than it did before, at 75mhz it got flakey but at 220 it would be 73.33mhz, and post back results. i doubt my memory will do much more than 225 at this voltage im afraid so more vdimm=good:toast:
I understand what you're saying. That's what I was getting at in one of my posts above.... the memory divider shouldn't have anything to do with the PCI/AGP lock, it's a separate issue. In theory, the lock either works or it doesn't, the divider only sets the memory speed relative to FSB. The AGP/PCI bus is still going to scale up with FSB, regardless of the divider. Of course, that doesn't mean there couldn't have been a quirk in the previous BIOS that kept the lock from working when a 1:1 ratio was selected.
So basically, you know that before you couldn't get past a certain point on FSB, regardless of setting your memory speed lower, indicating that your AGP bus was going too far out of spec for the ATI card. Now you can ramp all the way up to 240MHz, but since your memory has to run async at that FSB, you're seeing a big performance hit. That's interesting....I had read that A64's didn't take a big hit running async. But that might have been an nForce3 250 preview I was reading. Anyway, the lock must be working, because regardless of memory divider, your AGP bus should be 240/3 = 80, and it's unlikely your 9800XT would take kindly to that. My 9800XT pukes in the mid-70's.
I looked at some screen shots of the BIOS, and the divider nomenclature doesn't make any sense. I would normally say that at 200MHz FSB, memory is running at 400MHz, but that's DDR 200x2, so the ratio is really 1:1. The Asus BIOS says that 1:1 is 200 MHz DDR, and 2:1 is 400 MHz DDR. So what do you set to get stock PC3200 speeds at 200MHz FSB? 1:1 or 2:1 ? It it's 1:1, why would you even need to have a 2:1 option...who's got memory that's going to do 800MHz? If it's 2:1, why would you bother putting the 1:1 option...who's going to buy a Socket 939 CPU and then buy memory that can only do 200MHz DDR?
yes. to get 1:1 you have to set 2:1 in bios. i noticed that on start up when i set 1:1 and got 45s superpi runs, checked cpu-z and sure enough i was only runn'n 100mhz. To make it easy just take what they say, example 5:3 and just make it 5:3x(2) in your head. say the divider a:b and you have a 200mhz fsb. (200xa)/b would give you your actualy fsb speed.
Got it. Makes sense now. Well, not that it makes sense for them to do it that way, but at least I understand how to select the right value in the BIOS.
Thanks for all the testing tonight. I feel better now knowing that the lock should work once I get my board up and running.
then this should make you cream your panties like a 19 year old prom date(or was that just mine:D )
http://www.mechwarrior.tv/pictures/270HTT.jpg
and the link to go along with it(since it shows 200fsb in 3dmark)
http://service.futuremark.com/compare?2k1=7902373
Here's bandwidth running 224X12 @ 1:1. Timings are 2-3-2-5.
http://img13.photobucket.com/albums/...dra_224X12.jpg
Mongoose420,
Let the world see it, man!
Mongoose420@ 270fsb, 5:4
GJ!!
lol thanks. used to the FM forums 800x600 rules so i post links. Plus it makes the lazy of you actualy click a button:stick: :D
also here's pifast and super pi. not to shabby for a 2.4ghz processor i dont think
http://www.mechwarrior.tv/pictures/270HTTpfast.jpg
http://www.mechwarrior.tv/pictures/270HTTsuperpi.jpg
I'm using an R360 cored 9800 pro mate!Quote:
Originally posted by Mongoose420
you have a nvidia card legion? I tried 240x10 and a divider and 3dmark crashed right on video start up. R9800xt couldnt take that much agp bus i guess since i can bench at 220x11
Just off to try bigtoes bios will report back later!
Oh, we LIKE piccies here.... (FUGGER has lots of b/w :D ) and encourage them.... just a slight resize here.Quote:
Originally posted by Mongoose420
lol thanks. used to the FM forums 800x600 rules so i post links. Plus it makes the lazy of you actualy click a button:stick: :D
also here's pifast and super pi. not to shabby for a 2.4ghz processor i dont think
http://www.mechwarrior.tv/pictures/270HTTpfast.jpg
http://www.mechwarrior.tv/pictures/270HTTsuperpi.jpg
Mongoose420 PiFast @ 2.4gHz:
i cant wait till my v5 gets here. i'll match it up with my WW with a custom hold down adaptor and see what she's got. BTW i didnt test anything. figured out the most i could run HTT with my memory at 220 5/6 and went for it. i imagine 280 will fall also soon but im doing the forum shuffle with the locks news and proof(i dont think anyone can say at 270htt that the locks are not working)
and you're on 1005.20 right? from bigtoe?
yup. as soon as i was told the file need'd the .rom extension put on it i was flash'n away.
In fact i flashed use'n asus windows based flash menu that worked flawlessly btw(abits and DFI's both worked exelent for me as well tho).
im not prepared to go over 1.55v with the stock cooler. Wait till i get my v5 and keep a towl handy:banana:
well i seen what happens at 1.75v and the stock cooler allready with a bios bug. Lets just say i seen mid 60c temps and climbing very very fast as soon as i got into bios. it wasnt pretty. Given the fact ive allready not fried my new $500 dollar chip once im going to play it safe on the old vcore adjustments.
Does this happen to others? My A8V freezes on restart everytime after I lock the AGP/PCI freq. FYI, I am using Bigtoe's BIOS.
I don`t understand the dividers...I top out at 259fsb x 11 @ 1.8v with the 3:2 divider, 5:3 would slow it down somemore(I think) but it won`t boot at 259 @ 5:3.
Is there a BIOS that will increase the multiplyer?? 11 is the highest available.
I assume you mean replace the .20 with .rom?Quote:
Originally posted by Mongoose420
yup. as soon as i was told the file need'd the .rom extension put on it i was flash'n away.
At what FSB did you break even with the performance hit from running async, compared to your results at 215 1:1?
Same here mate!Quote:
Originally posted by H2OGun
Does this happen to others? My A8V freezes on restart everytime after I lock the AGP/PCI freq. FYI, I am using Bigtoe's BIOS.
I'm using a 3500.
Anyone know how to flash back using afudos as it's been years since I owned an asus board?
Since I am not at home at the moment and if I understand the conversation between stevenmh and mongoose correctly, can you try to set the FSB/memory ratio to anything other than 2:1 (DDR400) then lock the PCI/AGP freq again. Let's see if this still causes system freezes.Quote:
Originally posted by legion
Same here mate!
I'm using a 3500.
Anyone know how to flash back using afudos as it's been years since I owned an asus board?
Rgd your question about afudos, I think the command is something like this:
AFUDOS /iBIOS name
for example: if you have bios file called av81003.rom then you type "AFUDOS /iav81003.rom" but off course you have to boot in DOS mode in order to do this.
Make yourself a bootable floppy say with Dr DOS Disk For Bios Flashing over at http://www.bootdisk.com/Quote:
Originally posted by legion
Same here mate!
I'm using a 3500.
Anyone know how to flash back using afudos as it's been years since I owned an asus board?
Stick the flash utility and the BIOS on the floppy, and boot to it.
At the DOS prompt, type the command
A:\>afudos /i[filename] where filename is the BIOS file name incl the extension as in .ROM
sorta like http://www.svcheats.co.uk/adslfiles/afudos.jpg
That sounds similar to experiences from those who had to RMA their KV8 Pro boards when they failed to work with AGP/PCI lock with Abit's BIOS 14.Quote:
Originally posted by H2OGun
Does this happen to others? My A8V freezes on restart everytime after I lock the AGP/PCI freq. FYI, I am using Bigtoe's BIOS.
no dont replace .020 with .rom. add it on to the end. That file is just missing the .rom extension so tack it on at the end.Quote:
Originally posted by stevenmh
I assume you mean replace the .20 with .rom?
At what FSB did you break even with the performance hit from running async, compared to your results at 215 1:1?
Not to mine. ive run the 2:1 the 3:2 and the 5:3 dividers as there shown in the bios with no problems with the lock or freezing up:confused:Quote:
Does this happen to others? My A8V freezes on restart everytime after I lock the AGP/PCI freq. FYI, I am using Bigtoe's BIOS.
Just take how they look in bios and run them in this equasion to determin what divider you want to run A:B as a example. (FSB*A)/(B*2). So a 5:3 divider with a 260fsb would be (260x5)/(3x2)=216.666mhz. Your only runn'n your memory at 194mhz.:eek: It tops out there? ive checked with cpu-z and my take on the dividers seems to be correct. maby you should check with cpu-z or similar program to see exactly where your running.Quote:
I don`t understand the dividers...I top out at 259fsb x 11 @ 1.8v with the 3:2 divider, 5:3 would slow it down somemore(I think) but it won`t boot at 259 @ 5:3.
Is there a BIOS that will increase the multiplyer?? 11 is the highest available.
Now for NEW news....
Aparently hypertransport setting of 1000(LDT x5) does not do a goddamn thing. when set to 1000 you get a x4 LDT when you set it to 800 you get a x4 LDT.
http://www.mechwarrior.tv/pictures/SS/htt-stock.jpg stock HT set to 1000
http://www.mechwarrior.tv/pictures/SS/htt-test.jpg 270 with a 5:6 divider. HT set to 800
1000+ hypertransport seems be good to go but the bios setting is incorrect at least with the 1005.020 bios.:eek:
I get reboots with Bigtoes BIOS at stock speeds, halfway through N-Bench.
It`ll run 225fsb 1:1 and the bandwidth is higher then at 259 3:2
Makes sense. At 259 3:2, your memory is running under 200.Quote:
Originally posted by Rukee
It`ll run 225fsb 1:1 and the bandwidth is higher then at 259 3:2
I haven't had any reboot problems yet. Knock on wood. So far, I've gotten 233FSB @ 1:1 with timings of 2-3-2-5.
Has anyone done the v-core mod on this mobo yet?
OPP
Well, my board just died. I loaded big toes bios, no lock present. Was rebooting at 235 x 12.5 and came back from getting something to drink to the smell of burnt slicone. Doesn't look like anything else got fried, but cant check the cpu. It doesn't smell burnt. The smell is comming from a FET just under the agp slot. I placed an rma for the cpu just in case and will order a new board tomorrow.
The people who have working locks where did you order the board.
Hey there, pkrew. Sorry to hear about your board. I got mine at Zipzoomfly. The cheapest price I've seen is MWave at $148. Link
:banana: :banana: happen. I ordered mine from zzf as well. I'll overnight another one from them. They'll be getting this board back anyway, but I need to make sure the cpu is ok. Not sure what happened, no condensation, no mods on the board.
ouch. yea i got mine from zzf.com also and the locks working perfectly
Yep, well ordered it. Had it sent free 2nd air. $32 for overnight was just too much. I guess I'll have two of them at some point. Hopefully 1 will have working locks
Here's some info on my setup:
The beta BIOS does activate the lock positively.
My serial number is 45MGOJ2879.
My best results come off of 2:1 mem timing. Upping the HT and using a memory divider yields nothing worth chasing at this point.
My Corsair XLS doesn't mind 225mhz at 2.0-2-2-5 1T with only 2.8v.
Upping to 230mhz kills my ATI card. Locking APG/PCI at 2:1 mem divisor lets me run memory up with a slightly relaxed 2.0-3-2-5 1T (yes I know I dropped the trcd).
I find the board to be very stable. Too many BIOS changes made at the same time when you are near the fringe causes a non reboot to happen every now and then. A quick reset posts fine.
I've really only tinker for a few hours (had to work this weekend), but here is just some baseline stuff:
http://home.comcast.net/~mdzcpa/PI29.JPG
http://home.comcast.net/~mdzcpa/12925.JPG
Not bard for starters. I'll have more time this evening to tinker.
My serial number is only 7 away from your's, 2872, so hopefully the lock will work on mine also when I get my rma'd processor back.Quote:
Originally posted by mdzcpa
Here's some info on my setup:
The beta BIOS does activate the lock positively.
My serial number is 45MGOJ2879.
wow mine is 3279, much later..... hope it works
C
using the divider kills VC related benchs. :(
Who did you order your board from? Mine is probably one of the first from ZZF.Quote:
Originally posted by charlie
wow mine is 3279, much later..... hope it works
C
Hmmmmmm...
I seem to be finding random stability issues with the 1005 beta. Settings that are rock stable under 1003 are flaky under 1005 beta. I think they need to tweak the lock a little.
Otherwise I like the board. It may not be the best for the fringe guys, but it should make a fine 24/7 gamer.
I'll be shooting for 3Ghz+ and 13k+ in 2003 tonight.
What the heck?! Your 12V line is at 16.32V !!!!!Quote:
That's just a quirk in MBM :) It's fine under BIOS and Asus Probe.Quote:
Originally posted by SAE
What the heck?! Your 12V line is at 16.32V !!!!!
Good work mdzcpa! Hopefully Bigtoe will post another beta bios soon.
While I waiting for my new board to arrive. I just want to confirm the mods. Is this the vdimm mod and measuring point?
As long as the dividers are not working I wont try the vcore in case I want to rma. But just in case there are 2 zero ohm resistors. The one on the left is the one OPP pointed out and is probably correct. Also provided you could cut a leg that small. Could you just cut pin 6, if I remember correctly, of the chip and solder a resistor between the cut legs. Then you wouldn't need to remove a resistor
Here's a pic of the spot
Phew... Quite sure would be deadly ;)Quote:
Originally posted by mdzcpa
That's just a quirk in MBM :) It's fine under BIOS and Asus Probe.
is there a working windows based clock genorator for this board??
Well, this is the best I can do so far with BH-5 at 2-2-2-5 and watercooling. Bandwidth is very good, but I just need more VDimm. I'm very interested in that device Bigtoe has been showing off. May be just what I need. :D
http://img13.photobucket.com/albums/...us_desktop.jpg
That's it. Pins to ground screw.Quote:
Originally posted by pkrew
While I waiting for my new board to arrive. I just want to confirm the mods. Is this the vdimm mod and measuring point?
Have you tried to run asynch? i got better scores/times at 270x9 runn'n the 5:3 divider 221mhz ddr than i got at 220x11 1:1. about 400 points higher in 3dmark01se.Quote:
Originally posted by sierra_bound
Well, this is the best I can do so far with BH-5 at 2-2-2-5 and watercooling. Bandwidth is very good, but I just need more VDimm. I'm very interested in that device Bigtoe has been showing off. May be just what I need. :D
http://img13.photobucket.com/albums/...us_desktop.jpg
Also has anyone played with the asynclat thing? Or even knows what it does. google comes up with zilch.
async latency
lol i understand that. still doesnt tell me what it effectivly does.
lol. got to assume lower is better i guess.
Anyone here have Corsair XL and pushed 3v+ in them yet effectively?
Thanks xgman, I appreciate your help :toast:Quote:
Originally posted by xgman
That's it. Pins to ground screw.
I can run 270X9 with 5:3 divider, but bandwidth takes a big hit. I get better SuperPI and Pifast times running 1:1 with the tightest possible timings. My Pifast time is actually not that bad. I would rank about 34th on the Hexus list. FX's generally don't need to run at a high FSB (HTT). Most of the top FX/SuperPI times were achieved with less than 240 HTT. If I could get 230 running 2-2-2-5 and 1:1, I'd be a happy camper. I think that's possible with a BIOS upgrade or Bigtoe's VDimm device.Quote:
Originally posted by Mongoose420
Have you tried to run asynch? i got better scores/times at 270x9 runn'n the 5:3 divider 221mhz ddr than i got at 220x11 1:1. about 400 points higher in 3dmark01se.
Also has anyone played with the asynclat thing? Or even knows what it does. google comes up with zilch.
wierd. according to sisoft sandra im gett'n slightly better memory bandwidth at 270 5:3 than i get at 220 1:1 with only a 1mhz ddr mhz diffrence.
6467/6397 at 270 5:3 221mhz ddr 2-2-2-5-7-16
6403/6334 at 220 1:1 220mhz ddr 2-2-2-5-7-16
i can post ss's if you like.
I will need 3.04vdimm to get 230mhz out of my memory at these timings in dual channel. And i'll need more than 230 to get where im going want to be when i get my V5+Dtek WW on this baby:D . Also how much have you guys been play'n with these memory timings. come'n from a nf7-s 2.0 im not real familiar with all of them but 2-2-2-5-7. I dont know what value changes the Dram idle timer:confused: and i have no clue what the other about 4 values would even do if i changed em.
also whats this "bigtoe's device"? If it doesnt require me to solder i might be up for it:rolleyes:
Check out this thread.Quote:
Originally posted by Mongoose420
also whats this "bigtoe's device"? If it doesnt require me to solder i might be up for it:rolleyes:
http://homepage.ntlworld.com/a.leach80/guess.JPG
:slobber: very nice. i hope that doesnt need to be spliced into the atx line, its hard to tell but it looks like its just a extension.
no word on price yet tho. 3.1~3.2vdimm would make me very happy.
Bigtoe's Alpha version can do VDimm up to 4.2v. But I'd be happy with 3.1-3.4v. :D
only problem is i dont want mine to go any higher than 3.4~3.5. Know'n ive got 4.2vdimm on this little card would kill all of my hardware LOL. i have no self control, one of the reasons ive shy'd away from vmods:p: I think i may be gett'n some ozc 3700EB if its available in a month or so anyway:D
The regular vdim mod is easy. just clip on and go.
xgman, how are those XL sticks? Are they worth the hefty price tag?
i dunno, i think alot of people(including myself) got caught up in the XL craze. most review sites only get to about 245 or so on em with fairly loose timeings. I wish there were more reviews of the ocz 3700EB. anatechs review is almost to good to be true for my new set up. 262 with 3-2-3-6 timings at 2.85v:slobber: 260x11 1:1 in DC:eek: im really hope'n this chip goes that high, 2.5ghz on stock voltage tho seems to be a good tell of at least 2.7ghz i think and 260x10 would be no sweat.
Probably not. I haven't really tested them yet. Some guy reported 250 at 2-3-3-5 with 3.2v on a P4. Bigtoe says extra voltage doesn't help OCZ's version of them that much so the jury is still out. Quite frankly I wish I could use my registered CH sticks from the sk8v on this board. they were kick ass sticks with 3v.Quote:
Originally posted by sierra_bound
xgman, how are those XL sticks? Are they worth the hefty price tag?
Yes, despite all the bad-mouthing about registered RAM, the Corsair R & RE sticks are pretty good. I think I'm going to stay with BH-5 until I see definite proof that there's something better. I have another pair of BH-5 sticks (PC3200) that hasn't been used much. I may give them a try at some point.Quote:
Originally posted by xgman
Quite frankly I wish I could use my registered CH sticks from the sk8v on this board. they were kick ass sticks with 3v.
well i think you guys that have the FX and the 3800_ chips will keep the value of bh-5 up. 240~250 1:1 with a unlocked chip moderate moderate voltage(3.2~) and 5-2-2-2 timings is going to be as good if not better than what us multiplier limites guys get out of 260~270 at 3-2-3-5 timings. Bh-5 just isnt going to be good enough if these chips overclock well. even 245~250 may not be enough to max out high end aircooled or moderate watercooled chips being stuck with a 11 multi.:stick:
Good points, mongoose. I want to try higher VDimm with the BH-5 first, before moving on to something else. At only 221 HTT and 2-2-2-5, I still get around 6600 in Sandra. Imagine what I would get at 230 HTT or higher. ;)
Hi there
Im gonna upgrade to AMD64 FX-53 this summer.
My current RAM is OCZ PC4000 gold..
It dows 275fsb stable using 2.8v and 283fsb using 3.2v, and both are 1:1 and 3-4-4-8
Should I get another RAM or is this good ?
Im cooling with Vapochill LS.