The Official OCZ VX Thread - Strictly Test Data Only!!!

[Tek]

Bandwidth looks like T1 to me. Nice RAM you got there, I hope my VX will do the same and I am happy

[cuddles]

Looks like 2T to me. I get more than 4K at like 200 Mhz. If you're using all 4 DIMMs on S939 it forces 2T command rate, which it looks like you're using.

EDIT : If you are using 2x512, it says single channel. So if you're S754, then I guess that's pretty decent for single channel.

[trans am]

No, he has single channel motherboard, this looks perfect. It's 1T.

[st0nedpenguin]

Does anybody know why Sandra says "2CMD" when I am running CPC Enabled?

Sandra has always reported 2CMD for me no matter what setting on what motherboard, as long as it says 1T in the BIOS, I wouldn't worry about it.

[conrad.maranan]

BIG THANKS TO conrad.maranan FOR HOSTING THESE FOR ME

Glad to help you out, Alex. Your bandwidth is right on the money. You're running 1T, regardless of what Sandra says. Once you dial in everything with A64Tweaker, you can expect at least 50MB/s for bandwidth.

[Rabbi_NZ]

Cheers for the feedback everyone, I was curious if maybe I was running 2T and my memory wasn't as good as I thought... this is a great relief, thanks!

[Rabbi_NZ]

10x260 2-2-2-10-1T 3.3vdimm 1.75vcore ROCK SOLID 100% STABLE with 2x512 VX

Thanks again for the hosting Conrad, appreciated bro

[EDIT]
My results to date:

1st stick 7x250 3.3v 2-2-2-10 1T MemTest #5 10 loops no errors
2nd stick 7x250 3.3v 2-2-2-10 1T MemTest #5 10 loops no errors
2x512 7x240 3.2v 2-2-2-10 1T MemTest #5 40 loops no errors
2x512 7x245 3.2v 2-2-2-10 1T MemTest #5 20 loops no errors
2x512 7x250 3.2v 2-2-2-10 1T MemTest #5 20 loops no errors
2x512 7x250 3.2vdimm 1.55vcore 2-2-2-10 1T Prime Blend 1 hour & SuperPI 32MB no errors
2x512 10x250 3.2vdimm 1.61vcore 2-2-2-10 1T Prime Blend 8 hour no errors
2x512 9x260 3.2vdimm 1.55vcore 2-2-2-10 1T errors in MemTest #5 straight away
2x512 9x259 3.2vdimm 1.55vcore 2-2-2-10 1T errors in MemTest #5 straight away
2x512 9x260 3.3vdimm 1.55vcore 2-2-2-10 1T MemTest #5 3 loops no errors
2x512 9x258 3.2vdimm 1.55vcore 2-2-2-10 1T MemTest #5 6000-7000 errors in 400-500 loops
2x512 10x256 3.2vdimm 1.61vcore 2-2-2-10 1T SuperPI 32MB pass, Prime-Blend fail in 30mins
2x512 10x256 3.2vdimm 1.65vcore 2-2-2-10 1T Prime-Blend 11hours no errors
2x512 7x257 3.2vdimm 1.55vcore 2-2-2-10 1T MemTest #5 8 errors in 10 hours
2x512 10x260 3.3vdimm 1.65vcore 2-2-2-10 1T Prime errors within 30min / SuperPI 32MB fails
2x512 10x260 3.4vdimm 1.65vcore 2-2-2-10 1T Prime errors within 30min / SuperPI 32MB fails
2x512 10x260 3.4vdimm 1.71vcore 2-2-2-10 1T Prime Large FFT no errors in 1hour / SuperPI 32MB fails
2x512 10x260 3.4vdimm 1.75vcore 2-2-2-10 1T SuperPI 32MB Passed
2x512 10x260 3.3vdimm 1.75vcore 2-2-2-10 1T SuperPI 32MB Passed / Prime 4096-128k 10hrs+ stable / Sandra Mem Stable / MemTest stable

[aCidbAbY]

we have almost the same setup! what are your timeings?
i cant beat a 36/s pi score!

[LilGator]

"2x512 10x260 3.3vdimm 1.75vcore 2-2-2-10 1T SuperPI 32MB Passed / Prime 4096-128k 10hrs+ stable / Sandra Mem Stable / MemTest stable"

That's what I'm looking to do

awesome...

[Rabbi_NZ]

we have almost the same setup! what are your timeings?
i cant beat a 36/s pi score!

Keep in mind I have had this rig for a week total, it's my first AMD64 rig, and it's my first DFI... much more tuning to do!

Here they are:

Ratio 1:1
CPC Enabled
Tcl 2
Trcd 2
Tras 10
Trp 2
Trc 15 (will tighten once I find close to final clock)
Trfc 17 (will tighten once I find close to final clock)
Trrd 4 (may tighten once I find close to final clock)
Twr 3 (may tighten once I find close to final clock)
Twtr 2 (may tighten once I find close to final clock)
Trwt4 (may tighten once I find close to final clock)
Tref 3120 (may fiddle with this later too)
Twcl Auto
Skew Control Auto
Skew Value 100
*All the rest Auto

[hollywood]

Yo...someone PM me please as to where I can get this OCZ VX.

Thx...

[Playful_Buffalo]

Fry's has all the newshipments for the rest of this year (like 25 days )

[conrad.maranan]

Yesterday afternoon, I mentioned that I had a surprise. I've been wishy-washy about buying a new processor, especially since I'll be doing a very major upgrade in the near future. Anyway, my wife picked up my 3500+ from mWave while I was at work yesterday. This was the surprise I was talking about over here. I was mentally exhausted when I got home last night. I had a meeting to attend early this morning, but I took the rest of the day off so I can have some quality time with my new CPU. "Enough of the story telling, Conrad. Just get to the point, damn it!"



First of all, CPU code is as follows:

ADA3500DEP4AW
CBAXC 0437SPBW
1076514I40252

Week 37? Big deal, nothing new, who cares? The first thing that I noticed was that the CPU voltage set in BIOS was the exact same voltage registered with CPU-Z. I tested accuracy with a digital multimeter and the voltage was true. BIOS is still 1.36 Beta. I then made the necessary adjustments to have my CPU and memory run 1:1 at 10x250MHz. Guess what? It worked! At this point, I was jumping for joy because I did not have to drop my CPU multiplier down to a mediocre value to push my memory at this frequency. Moving on, I dialed in my DDR Booster to a digital multimeter-verified 3.51V. After several hours of testing, the images below depict the best I was able to pull off.







Here's what's interesting: I was able to do 265MHz Memtest and Super PI 32M stable on the two 3000+ Winchesters that I tested with previously. For some reason, the magic number is not making itself known with this CPU. VDIMM has been adjusted in 0.05V increments from 3.3V all the way up to 3.8V. 265MHz is just not manifesting itself right now. I haven't experimented with lower CPU multipliers yet, but I'm beyond happy with where I'm at considering I could not do this with the prior Winchesters tested.

Maximum CPU speed has not yet been tested, so I'm hoping for a few more megahertz later on this evening. Current BIOS setting for CPU voltage is 1.550V + 5% using AMD's retail-boxed HSF. Once I find my maximum CPU frequency, I'll experiment with memory dividers to discover my sweet spot.

Last but not least, I ripped the heatspreaders off of my VX modules. Yes, I'm aware that I voided the warranty by doing so, but the warranty has been voided since day one considering I pushed up 3.8VDIMM on the first night. What did I gain by doing this? Max stable Memtest was 255MHz with the heatspreaders intact. A 2MHz gain isn't anything extreme. It doesn't even justify voiding a lifetime warranty. But I'll do anything for an increase in speed, even if it's just 2MHz. I'm just xtreme like that.

Cheers!

[bacon]

glad your enjoying your newcastle mate, i get my second one tomorrow, as well as the stupid company who delivered the old one picking the old one up for my refund (as meant to be a wincehster as they charged me £20 over NC price)

So hopefully its a nice clocker unlike this one i have at the min

lol 2mhz and no warranty shame on you, stick em back on, they'll never know

[Rabbi_NZ]

Nice results conrad Sounds like your MemController is maxin' out... drop a few multis, raise VCore a bit, and then see if you can push those VX's a bit higher

[conrad.maranan]

Nice results conrad Sounds like your MemController is maxin' out... drop a few multis, raise VCore a bit, and then see if you can push those VX's a bit higher

I'm workin' on it.


On a side note, I'm experiencing the same problem trans am was with regard to the memory dividers. The only difference is he wasn't able to use any divider and I'm able to use only the 1/2 (100MHz) divider. Strange.

I'll swap in my Platinum Deuce modules and see if it's the same deal for that RAM.

[conrad.maranan]

Alright, folks. I thought this was interesting. Draw your own conclusions as I'm close to finished formulating mine.

Maximum CPU Speed: 2730MHz (10x273MHz) | Memory 1:1 | OCZ Platinum Deuce
Maximum CPU Speed: 2650MHz (10x265MHz) | Memory 1:2 | OCZ VX
Maximum Memory Frequency: 280MHz | 2.5-3-3-10 | 2.84VDIMM | OCZ Platinum Deuce
Maximum Memory Frequency: ???MHz | 2.0-2-2-10 | ?.??VDIMM | OCZ VX

After I test for maximum memory speed on the VX modules, I'm plugging the Platinums back in to find out what I can do at 2.5-4-3-10 and 2.5-4-4-10. I'm just having too much fun today.

[conrad.maranan]

There was no increase in frequency to be realized by lowering the CPU multiplier. I went down the line until I hit the 5 multiplier. 5, 6, 7, 8, 9, and 10 all yielded the same maximum memory speed of 257MHz.

And then I decided to bump it up a notch to 258MHz. A few hours ago, this frequency would error out on Test #5 at the second loop. Perhaps I just have to burn the new 3500+ in for a few days. Check out the differences 1MHz resulted in.







I'll start looping 259MHz now and see how that turns out.

[bacon]

is your cpuz out of date or something as its showing as a newcastle

anyway nice so far dude, you preferring it to the old chip? and ya linking the VX loving mine, gotta wait for my chip now as it hasnt come today like it should have

[trans am]

Actually, it's a 3500+ Winchester. Let's see how long this one lasts me.

First of all, CPU code is as follows:

ADA3500DEP4AW
CBAXC 0437SPBW
1076514I40252

I'm workin' on it.

AW=CG = in this case "NEWCASTLE"

On a side note, I'm experiencing the same problem trans am was with regard to the memory dividers. The only difference is he wasn't able to use any divider and I'm able to use only the 1/2 (100MHz) divider. Strange.
I'll swap in my Platinum Deuce modules and see if it's the same deal for that RAM.

I never bothered with the 100 divider, but it's strange that 133 and 166 don't work at all from the bios, but a64 tweaker works. However, I am able to run dividers in the bios with my el4200 plat. and the PDP xbl. OCZ guys please make a note.

[Nanobot]

is your cpuz out of date or something as its showing as a newcastle

That's because conrad's CPU is a Newcastle...

Here's a nice interactive OPN utility form ocinside.de

Use it so you dont suffer foot-in mouth desease again .

Just punch in conrad's AMD 64's code :ADA3500DEP4AW,you'll see.

[trans am]

That's because conrad's CPU is a Newcastle...

Here's a nice interactive OPN utility form ocinside.de

Use it so you dont suffer foot-in mouth desease again .

Just punch in conrad's AMD 64's code :ADA3500DEP4AW,you'll see.

I thought the whole point of conrad buying the cpu was because it was a newcastle and woun't have the sub zero bug and less likely to have issues with Prime95 etc. He is getting a phase unit soon, so this would make sense that he bought this intentionally. Conrad, why did you say it's a WInchester?


On a side note, I got my replacement VX yesterday. I didn't bother playing with it because of my current mobo/cpu problem. I'm going to try these over at jikdok's place tonight.

[bacon]

Actually, it's a 3500+ Winchester. Let's see how long this one lasts me.

I'm workin' on it.

errm ok fair enough, i thought he had a winchester for this reason

[bacon]

Use it so you dont suffer foot-in mouth desease again .

Just punch in conrad's AMD 64's code :ADA3500DEP4AW,you'll see.

i wasnt wrong from what he personally typed

i know what codes a newcastle is :

[conrad.maranan]

I'm sorry for all of the confusion. My mistake, and boy, I sure have been making a lot of mistakes lately. It is a Newcastle (AW) 3500+. Anyway, no Prime issues here.