Has anyone else tried both the Rev1 and Rev2 965P-DS3 boards? With the now-dead Rev1, I could get 3.2ghz at 1.35v, and 3.3ghz at 1.375v. The new Rev2, with the same settings, same F7 Bios...takes 1.36v to run at 3.2ghz, and anything over 3.2 (even a few mhz at 1.375v) fails Orthos within a minute. Are there any tricks for this new revision board that I don't know about or is it just a step backward in terms of overclockability?
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Navigation
Install the app
How to install the app on iOS
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
More options
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
DS3 Rev1 vs Rev2 overclocks
- Thread starter ATW
- Start date
Haha glad you went the other way that I did. I have a now dead ds3 rev2, and just got a used rev1 c1.
Things I've noticed:
With the rev2, I also had to run 1.375v to get 3.2 stable on my e6400. But, 1.375v set in bios was 1.328v idle and 1.296-1.312v load. So droop isn't too bad, but overall voltage is way lower than set. Setting the same 1.375v with my rev1, I see around 1.36-37 idle and 1.34-36 load. Neither droop nor undervoltage is a big issue here at all. I didn't try setting vcore lower since I'm trying to focus on 8x450=3600.
With the rev2, it seemed that 1.6v was unstable regardless. I'm not exactly sure if it was just my board or some other parameter, but the max usage vcore was the last 1.59v. So with my rev1, I never bothered with 1.6v. With rev2, 1.59v set resulted in 1.52-54v idle and 1.48-50 load. This was particularly problematic for me since 1.48 will eventually cause it to fail orthos with one of my e6400's. With my rev1, 1.59v results in 1.57v idle and 1.52-54 load.
Sorry, what I said was definitely way more than you needed to hear. Check your vcore with cpuz or new beta speedfan or et5. cpuz is accurate in voltage until you get into the 1.4's somewhere and then it drops to static 1.213.
Also, it seems that with the rev2 I had, vdimm was set 0.1v higher than whatever you choose in the bios. I was seeing 2.3v when I set +0.4v. With my rev1, I see 0.1v lower!! This caused some headaches for a long time. Setting +0.4v resulted in 2.2v. This can be measured with a dmm (easy to do on rev1, I never looked to figure out how to do it on rev2, but they changed all the electrical components to something new and there's no traditional inductor we're used to seeing). However, you can use the new beta speedfan (or I suppose a new mbm) to look at vcore2, which would actually be vdimm. I've found that the reported vcore2 (vdimm) is about 0.02v lower than what is set, but it could mean something at the same time.
Regardless, as for rev2 boards, I had a lot better luck with mine than I am this used rev1. I have d9gmh, or am led to believe so. With rev2, I could run 400 @ 3-3-3-9 and 500 @ 4-4-4-12 and 2.3v actual voltage. Never tried if lower volts would work also or if I could get timings down further, but I was happy with those numbers. Now with my rev1, it gets extremely sensitive about ram, and often times, I'll have to leave the computer off for a few minutes before it'll want to post again (even if I reset the cmos). And sometimes I've had to pull a stick of ram for it to even post the first time. But then when I put it back again, it works. I haven't been able to run 3-3-3-9 at 400 anymore and can't do 4-4-4-12 at 500 either. In fact, currently I'm at 450, and still can't do 4-4-4-12. So I dropped voltage down a bit since there wasn't any point.
So all in all, my new used rev1 c1 is extremely tempermental, and I think there might actually be something wrong with it (haven't confirmed anything yet but it would explain why it was returned and resold as openbox). I don't even dare fiddle with my bios anymore as long as it's working as it is currently, even though I'd like to see better numbers, just because the last time I tried, it took me about an hour and a half to coax the board to accept the settings I'm using back.
The rev2 seemed to have less of a cold boot issue. It still did, and very annoying it was. But it never left me needing to yank ram for it to post or constantly having to reset cmos, or have to wait half an hour before it'll post again. Also, I was able to do 7x500 with the rev2. I'm afraid to try with this one, although I'm really tempted to. I just really don't want to have to sit and wait an hour to see a post again.
Things I've noticed:
With the rev2, I also had to run 1.375v to get 3.2 stable on my e6400. But, 1.375v set in bios was 1.328v idle and 1.296-1.312v load. So droop isn't too bad, but overall voltage is way lower than set. Setting the same 1.375v with my rev1, I see around 1.36-37 idle and 1.34-36 load. Neither droop nor undervoltage is a big issue here at all. I didn't try setting vcore lower since I'm trying to focus on 8x450=3600.
With the rev2, it seemed that 1.6v was unstable regardless. I'm not exactly sure if it was just my board or some other parameter, but the max usage vcore was the last 1.59v. So with my rev1, I never bothered with 1.6v. With rev2, 1.59v set resulted in 1.52-54v idle and 1.48-50 load. This was particularly problematic for me since 1.48 will eventually cause it to fail orthos with one of my e6400's. With my rev1, 1.59v results in 1.57v idle and 1.52-54 load.
Sorry, what I said was definitely way more than you needed to hear. Check your vcore with cpuz or new beta speedfan or et5. cpuz is accurate in voltage until you get into the 1.4's somewhere and then it drops to static 1.213.
Also, it seems that with the rev2 I had, vdimm was set 0.1v higher than whatever you choose in the bios. I was seeing 2.3v when I set +0.4v. With my rev1, I see 0.1v lower!! This caused some headaches for a long time. Setting +0.4v resulted in 2.2v. This can be measured with a dmm (easy to do on rev1, I never looked to figure out how to do it on rev2, but they changed all the electrical components to something new and there's no traditional inductor we're used to seeing). However, you can use the new beta speedfan (or I suppose a new mbm) to look at vcore2, which would actually be vdimm. I've found that the reported vcore2 (vdimm) is about 0.02v lower than what is set, but it could mean something at the same time.
Regardless, as for rev2 boards, I had a lot better luck with mine than I am this used rev1. I have d9gmh, or am led to believe so. With rev2, I could run 400 @ 3-3-3-9 and 500 @ 4-4-4-12 and 2.3v actual voltage. Never tried if lower volts would work also or if I could get timings down further, but I was happy with those numbers. Now with my rev1, it gets extremely sensitive about ram, and often times, I'll have to leave the computer off for a few minutes before it'll want to post again (even if I reset the cmos). And sometimes I've had to pull a stick of ram for it to even post the first time. But then when I put it back again, it works. I haven't been able to run 3-3-3-9 at 400 anymore and can't do 4-4-4-12 at 500 either. In fact, currently I'm at 450, and still can't do 4-4-4-12. So I dropped voltage down a bit since there wasn't any point.
So all in all, my new used rev1 c1 is extremely tempermental, and I think there might actually be something wrong with it (haven't confirmed anything yet but it would explain why it was returned and resold as openbox). I don't even dare fiddle with my bios anymore as long as it's working as it is currently, even though I'd like to see better numbers, just because the last time I tried, it took me about an hour and a half to coax the board to accept the settings I'm using back.
The rev2 seemed to have less of a cold boot issue. It still did, and very annoying it was. But it never left me needing to yank ram for it to post or constantly having to reset cmos, or have to wait half an hour before it'll post again. Also, I was able to do 7x500 with the rev2. I'm afraid to try with this one, although I'm really tempted to. I just really don't want to have to sit and wait an hour to see a post again.
Is there really that much droop on these boards? I've never believed software when it comes to voltage readings, so I just go by whatever I have it set to. The temps seem to be jumping up at about the same voltage on both boards, so the amount of droop can't be much different between them. I'm running Orthos right now at 1.368v and the CPU-Z reading is 1.296v.
WickedWeasel
Limp Gawd
- Joined
- Jul 13, 2004
- Messages
- 444
Isn't Vdroop on the DS3 board lower than on the other P965 boards?