What's a safe voltage for DIMM stability.

Status
Not open for further replies.
Joined
Sep 25, 2003
Messages
8
Reaction score
0
I'm running my Mushkin Black Level II PC3500 (rated 2-2-2-5) at normal stock FSB 200MHz. I'm using SPD timings (2-2-2-5) and when changing the performance to anything but auto (ie. Turbo, Street Racer, F1) I get a FATAL ERROR in prime95.

MemTest checks out okay after hours of testing, but not so with Prime95. After the third line of prime95 I get a FATAL ERROR message.

So I'm thinking of upping the DDR voltage instead of sending the memory right back to Newegg. Any ideas how high I can safely up the voltage to acheive stability? Thanks in advance.

S.Anderson
 
What CPU and motherboard are you using...?

What are the settings for the multiplier for the CPU and is the memory in sync with the FSB or not...?

Also is your Vcore set as std..? And what is the VDD set at now...?
 
On Fri, 03 Oct 2003 12:55:04 GMT, "Rieper" wrote:

I'm running my Mushkin Black Level II PC3500 (rated 2-2-2-5) at normal stock FSB 200MHz. I'm using SPD timings (2-2-2-5) and when changing the performance to anything but auto (ie. Turbo, Street Racer, F1) I get a FATAL ERROR in prime95.

MemTest checks out okay after hours of testing, but not so with Prime95. After the third line of prime95 I get a FATAL ERROR message.

So I'm thinking of upping the DDR voltage instead of sending the memory right back to Newegg. Any ideas how high I can safely up the voltage to acheive stability? Thanks in advance.
2.8 volt should be safe for DDR ram (default 2.5). From 3.0 and above I would be afraid of melting.

It can also be helpfull to raise the chipset voltage a bit, 0.1 to 0.2 volt. But try without first.

best regards

John
 
Pentium 4 3.20c (hyperthreading enabled)
Mobo: Abit IS7 (regular version)
VCORE is the default setting, I don't need to overclock my stock P4 3.2GHz VDD is also at the default setting. Haven't really tweaked anything.

I think I might try 2.8v and see what goes. Wish me luck!

"Morgan" wrote:

What CPU and motherboard are you using...?

What are the settings for the multiplier for the CPU and is the memory in sync with the FSB or not...?

Also is your Vcore set as std..? And what is the VDD set at now...?

--
Regards

Morgan

...
 
By the way, the Mushkin is 2x512MB sticks of DDR PC3500.

"Rieper" wrote:

Pentium 4 3.20c (hyperthreading enabled)
Mobo: Abit IS7 (regular version)
VCORE is the default setting, I don't need to overclock my stock P4 3.2GHz VDD is also at the default setting. Haven't really tweaked anything.

I think I might try 2.8v and see what goes. Wish me luck!

"Morgan" wrote:

What CPU and motherboard are you using...?

What are the settings for the multiplier for the CPU and is the memory in sync with the FSB or not...?

Also is your Vcore set as std..? And what is the VDD set at now...?

--
Regards

Morgan

My noisy drive is now fixed....
http://www.flyinglizard.freeserve.co.uk

...
 
2.7 or 2.8 volts has been reported by many here in the group. do you have problems in any other applications in everyday use? if not, then Prime95 might not be an issue... I realize that you don't want instability lurking in the background...
 
Yes, as a matter of fact, when playing Battlefield 1942, for the first time ever, I got booted back to my desktop. I am not overclocking my Radeon 9800 Pro at all. So I'm pretty sure its the memory timings which are screwing me up.

After upping the voltage to 2.8, and retrying prime95, I still get the FATAL ERROR message. I don't even bother playing Battlefield 1942, I'm pretty sure it will crash.

Long story short, this memory is going back. I will stick with my PC3200 Mushkin (2x256MB) which runs for hours in prime95 and gaming at 2-2-2-5 (Turbo setting).

Thanks all. Very informative group for Abit customers. Will continue coming back and add my own input in the future.

S.Anderson

"TomG" wrote:

2.7 or 2.8 volts has been reported by many here in the group. do you have problems in any other applications in everyday use? if not, then Prime95 might not be an issue... I realize that you don't want instability lurking in the background...

--

Thomas Geery
Network+ certified

ftp://geerynet.d2g.com
ftp://68.98.180.8 Abit Mirror <----- Cable modem IP
This IP is dynamic so it *could* change!...
over 117,000 FTP users served!
^^^^^^^
...
 
You have had some good suggestions already but where do you see that the memory that you have bought is rated at 5.2.2.2

I must be looking at the wrong item as I see....

http://tinyurl.com/plfo

and written......
The CAS latency of this memory depends on the chipset you are using.

For Intel 875 & 865 chipsets, the rated specs are:
CAS latency: 2
RAS to CAS Delay: 3
Precharge Delay: 2
Active to Precharge Delay: 6
DDR Voltage 2.5-2.75V

Please note that this limitation is due to the chipset rather than the memory. Some customers may be able to achieve 2-2-2-6 CAS latency on the 875 and 865 chipsets, but we cannot guarantee this.
 
Did you read this item about Abit GAT settings on the Mushkin support page?
----------------------------
USER TIPS: enabling any of the different performance options will override the manual latency and frequency ratio settings and force a 1:1 CPU:DRAM ratio at 2:x:x:x latency settings, with the "Ultra" (now "F1") setting forcing a 2:2:2:5 latency configuration at DDR400 frequency.

We also found that using the "Turbo" (or "Street Racer") setting at 2:2:2:7 yields overall better performance since a longer tRAS prevents the controller from inadvertently terminating an ongoing data burst. Enabling the CPC (Command Per Clock) adds another small notch to the overall performance but can cause stability problems as well. The Read Delay timer should be set to 5-6 cycles for best stability and performance or else left on Auto.

We found that DIMMs that require CAS latencies of 2.5 or higher at DDR400 will not function in any of the available performance modes. The workaround is to set the performance tab to "Disabled". Even though this appears counterintuitive at first, it makes sense in that there is no point in reducing chipset latencies when the memory itself would add those latencies again at the back-end of the "food chain". The HW monitor shows voltage values that are below the real voltages, so better not crank them up too much.

Keep in mind that these recommendations are based on "snapshots" of a few of the current BIOS revisions and may change with future BIOS versions.
----------------

With Buffalo Tech PC3700 on an Abut IC7, I've got my ram set to the best timings I can manage, 2.3.2.6 @ 220mhz running 5:4. That gives me a 275 fsb and takes my P4C 2.4 to 3.3g. All benches, including memory, are right where they ought to be.

With the GAT settings at anything but auto, the system becomes unstable. On auto, it will loop MemTest and Prime95 without a hiccup at the above settings.

I've pretty much come to the conclusion that GAT is for those running 1:1 who don't know how to tweak their own memory settings. If you're overclocking at 5:4 or using higher latencies for a higher fsb, GAT will do more harm than good.

In any case, it seems to be a real crap shoot, working for some and causing serious problems for others.

The biggest frustration is that Abit has yet to reply, so far as I know, to any of the many requests from reviewers and customers wanting to know what those settings actually DO.

"Rieper" wrote:

I'm running my Mushkin Black Level II PC3500 (rated 2-2-2-5) at normal
stock
FSB 200MHz. I'm using SPD timings (2-2-2-5) and when changing the performance to anything but auto (ie. Turbo, Street Racer, F1) I get a
FATAL
ERROR in prime95.

MemTest checks out okay after hours of testing, but not so with Prime95. After the third line of prime95 I get a FATAL ERROR message.

So I'm thinking of upping the DDR voltage instead of sending the memory right back to Newegg. Any ideas how high I can safely up the voltage to acheive stability? Thanks in advance.

S.Anderson
 
On Fri, 03 Oct 2003 15:37:21 GMT, "Skid"
wrote:

Did you read this item about Abit GAT settings on the Mushkin support page?
----------------------------
USER TIPS: enabling any of the different performance options will override the manual latency and frequency ratio settings and force a 1:1 CPU:DRAM ratio at 2:x:x:x latency settings, with the "Ultra" (now "F1") setting forcing a 2:2:2:5 latency configuration at DDR400 frequency.

We also found that using the "Turbo" (or "Street Racer") setting at 2:2:2:7 yields overall better performance since a longer tRAS prevents the controller from inadvertently terminating an ongoing data burst. Enabling the CPC (Command Per Clock) adds another small notch to the overall performance but can cause stability problems as well. The Read Delay timer ...

Errrr.............. yes.

Posted by me to the Abit forums and in this newsgroup months ago and immediately afterI saw PAT (now GAT ) on the BIOS screen with no functional description. in the IC7-G manual. Fat lot of good that did............!

Oh, well ! What do you expect from a company with manual writers and engineers to whom English is a second language. Also, the lack of decent technical information in the any of the Abit MB manuals contributes to stable company employment. Keeps the bodies in Customer Support occupied.

Having written lots of technical documentation in my career, the Abit nonchalence makes my teeth grind...........

John Lewis

"Rieper" wrote:

I'm running my Mushkin Black Level II PC3500 (rated 2-2-2-5) at normal stock
FSB 200MHz. I'm using SPD timings (2-2-2-5) and when changing the
performance to anything but auto (ie. Turbo, Street Racer, F1) I get a FATAL
ERROR in prime95.

MemTest checks out okay after hours of testing, but not so with Prime95.
After the third line of prime95 I get a FATAL ERROR message.

...
 
"John Lewis" wrote:

Posted by me to the Abit forums and in this newsgroup months ago and immediately afterI saw PAT (now GAT ) on the BIOS screen with no functional description. in the IC7-G manual. Fat lot of good that did............!

Oh, well ! What do you expect from a company with manual writers and engineers to whom English is a second language. Also, the lack of decent technical information in the any of the Abit MB manuals contributes to stable company employment. Keeps the bodies in Customer Support occupied.

...

I agree with all but the last part. I'm not sure they DO employ anyone in customer support, preferring to fob it off on the vendors.

Abit makes terrific hardware and gives overclockers lots of tools to squeeze every drop of speed out of our systems.

But then we are forced to scrounge third-party websites and apply deductive reasoning based on mass experience to figure out what in the hell all those bios settings do.

GAT is the worst example yet, IMHO. I understand that they don't want to offend Intel by going into specifics on exactly how PAT is enabled on a chipset that doesn't officially support it.

What I can't understand is why they won't at least provide details -- even when asked directly in their own forums -- about something as simple as what ram timings change from Turbo to F1.
 
Status
Not open for further replies.

Latest threads

Back
Top