Bought OCZ EL PC3200 Platinum Revision 2, VERY UNHAPPY! Won't OC for BEANS!!!

Joined
Oct 17, 2004
Messages
31
Seriously,

After reading over a dozen reviews, all able to push this RAM really far with great timings, I decided it was worth the extra money (which I don't have) to get them. But for some reason, they won't OC.

One reviewer was able to get them to 263 @ 2.85V w 2.5-3-3-6

Another was able to get them to post at 275 @ 2.85V w 3-4-4-8

Many were able to get them up to 275, and some even GUARANTEED 250.

I can't get mine to post at 240 @ 2.9V w 3-4-4-10, and 250 is completely out of the question, no matter what I do.

So, lets see, 275 with low timings, compared to my WON'T EVEN POST AT 240 with WORSE TIMINGS and MORE VOLTAGE.

I'm very upset, and I'm the only person I can find online who got "duds", so I really hope it's not that. So...

Please, PLEASE, someone tell me what I'm doing wrong.

I have an Athlon XP-M 2500+, ABit NF7-S V2.0 (newest BIOS), 6600GT AGP.

Again, please help.

Thanks,

lllsynthtaxlll
 
These only start to shine after about 3.1 vdimm... tough luck there.
 
That doesn't really mean anything when others are getting great results at 2.85V.

In fact, all the reviews I read (that I remember) never pushed the voltage above 2.9 and got those great speeds, so what is the relevance of you telling me that they only shine after 3.1?

lllsynthtaxlll
 
those were old, "431" TCCD; the new "432" chips like higher voltage to achieve about the same results.
 
I have the OLD first week they came out still new in package (been 4+ months, paid 360.00 lol at newegg) :rolleyes: still waiting for a motherboard hahahah =(
 
No, most of the reviews were on Intel systems.

You're saying that the ones I got are different from the ones that were reviewed. So how come my RAM will actually work at 220 FSB with 2-2-2-5, but above that and it's like hitting a brick wall.

Where can I read more about this TCCD 431, 432 issue you mentioned? Is there a way I can be sure that's what I have?

Thanks,

lllsynthtaxlll
 
Personally, i think it's your motherboard holding you back.

I have really good RAM too, and when i attempted the overclock, i reached a 221 MAX (with low timings), even with my multiplier dropped.

If you had the DFI LanParty however, that's a different story. That board does some insane stuff with good RAM.

I may be wrong, but that's my opinion on the NF7-S.
 
Look at the chips. Careful with the sticker, though, you don't want to make it have wrinkles (b/c when you glue it back on after, it won't look convincing). I suggest peeling it off with a hobby knife. Also, I understand, OCZ uses chromed heatspreaders on that ram; it's also pretty easy to take off the clips without making any visible scratches (there will be scratches UNDER the clips though, but who can see?)

On top of the chip, above "TCCD" it will either say "431" or "432". That's how you know for sure :)
here's an example:
tccd2.jpg


edit: by the way - I know nothing about NF2 motherboars (since I never had ANY interest whatsoever in Socket A rigs), so that might be the issue as well.
 
lllsynthtaxlll said:
No, most of the reviews were on Intel systems.

Intel motherboards achieve higher FSB than nForce2, if you want the highest FSB possible on an nForce2 motherboard do what the above post suggest and get a DFI Lanparty.
 
You're not the only one.

I got a mobile 2400+ on a NF7-S 2.0 with the newest BIOS, and on 2.9v I can only get it stable at 214FSB 2-2-3-9(I keep it at 9 because I get the most performance with that, I can get it at 5). I don't know why, but my processor doesn't like anything higher then 2.4GHz, won't post at 2.5. Now I'm thinking of getting that OCZ DDR Booster..is it really worth it?
 
Use prime95 or sanda memory burn in. Let it run for a few days. It might overclock better. I had pdp the xblk version, which is the same ram chips and pcb as yours, but cheaper ;-). At first, it would not even post at 240mhz at any timings or voltage increase, but I ran it underclocked and at 2.65V for a few days at tight timings for a few days, and when I started tweaking my system again, it was able to boot. Plus, I can up the voltage to 2.85V, and get slightly better timings.
 
So what's with the DFI boards that make them so great at overclocking?

I really thought the NF7-S was the one to get for overclocking, but I never looked into the DFI boards, and I already spent to much money on what I did get.

Well, I might take off the heatspreaders tonight to check and see, but either way I'm very dissapointed.

Why is it that Intel can overclock their FSB so much more than AMD?

lllsynthtaxlll
 
Your motherboard is holding you back, nf7's only hit about 240-250 tops, thats with modded bios as well. and tccd doesnt need high voltage, 3v or less is good for about 275-300MHz if you got a good stick. HOWEVER there is supposed to be a new version of tccd that likes voltage and can do tight timings at high speeds with high voltage( 2-X-X-X). You should be able to hit lower latency at 240, so I would try to tweak a little. Other than that your board is maxxed out FSB wise. Not like it matters, 240, is a very respectable fsb for a athlon xp. You could go infinity or lanparty b and they seem to hit about 250-270 if you are lucky. However they have some problems last I checked, a new bios might have been released though fixing these problems.
 
Tccd 431 won't be effective with voltage above 2.7V. It will do 2.5-3-3-7 at 290 depends on how stable your system is and how your psu perform under heavy load. Here's mine using the same chip as yours with neo2
superpi3700.jpg
 
Well, thanks for all your help, but I'm just p*ssed that I spent the extra $$ on this RAM, and was basically mislead. I knew that there was a chance since all the reviews were on Intel that I wouldn't get as good an overclock, but not like this, this is ridiculous. I could have saved a lot of money getting some different sticks that would have performed the same for me.

And if I have 432, then it really p*sses me off when companies release a product that's great, then after it's been reviewed, puts in inferior parts.

I might spend some more time tweaking, but in the end I'm not getting anything close to what I expected, and would warn against buying this RAM if you have an Athlon XP system.

lllsynthtaxlll
 
lllsynthtaxlll said:
Well, thanks for all your help, but I'm just p*ssed that I spent the extra $$ on this RAM, and was basically mislead. I knew that there was a chance since all the reviews were on Intel that I wouldn't get as good an overclock, but not like this, this is ridiculous. I could have saved a lot of money getting some different sticks that would have performed the same for me.

And if I have 432, then it really p*sses me off when companies release a product that's great, then after it's been reviewed, puts in inferior parts.

I might spend some more time tweaking, but in the end I'm not getting anything close to what I expected, and would warn against buying this RAM if you have an Athlon XP system.

lllsynthtaxlll
Theres no different in term of performance and oc beetween 431 and 432. Its only week of production. Week 31/04 and 32 batches.
 
centvalny,

That's pretty impressive what you're running it, but I'm pretty sure A64 would have a much better overclock than AXP. Hopefully when I upgrade to A64, I'll still be able to use this RAM.

What RAM are you using?

Are you saying that if I have TCCD, I should try working at a lower voltage instead of always trying to get it at 2.9V?

lllsynthtaxlll
 
lllsynthtaxlll said:
Well, thanks for all your help, but I'm just p*ssed that I spent the extra $$ on this RAM, and was basically mislead. I knew that there was a chance since all the reviews were on Intel that I wouldn't get as good an overclock, but not like this, this is ridiculous. I could have saved a lot of money getting some different sticks that would have performed the same for me.

And if I have 432, then it really p*sses me off when companies release a product that's great, then after it's been reviewed, puts in inferior parts.

I might spend some more time tweaking, but in the end I'm not getting anything close to what I expected, and would warn against buying this RAM if you have an Athlon XP system.

lllsynthtaxlll

dude, your ram is running at 240MHz, stock is 200! Not bad to me. Also you should have researched how well your board overclocks fsb wise before expecting a 270Mhz fsb. This is how companies get bad reps
:rolleyes:
 
lllsynthtaxlll said:
centvalny,

That's pretty impressive what you're running it, but I'm pretty sure A64 would have a much better overclock than AXP. Hopefully when I upgrade to A64, I'll still be able to use this RAM.

What RAM are you using?

Are you saying that if I have TCCD, I should try working at a lower voltage instead of always trying to get it at 2.9V?

lllsynthtaxlll

dude, you always gotta teak and see what works best. Never try something and if it doesnt work post on the forums and say it doesnt work. You gotta tweak your heart out when it comes to overclocking and you need to do alot of very long testing. I wish you good luck :D and also since you seem to be new to overclocking I reccomend prime 95 to test your overclock stability (ram, fsb, and cpu).
 
lllsynthtaxlll said:
centvalny,

That's pretty impressive what you're running it, but I'm pretty sure A64 would have a much better overclock than AXP. Hopefully when I upgrade to A64, I'll still be able to use this RAM.

What RAM are you using?

Are you saying that if I have TCCD, I should try working at a lower voltage instead of always trying to get it at 2.9V?

lllsynthtaxlll
Yes, it will help. This is the same memory used (tccd431)on AXP with stock voltage, I believe it is 2.6V and xp2800 tbredB (2.25Ghz stock) and NF7S ver2
sandracpuid2800.jpg
 
I'm not sure where you got the impression that I was able to POST at 240, I wasn't.

So far the top I can get is 225 @ 2.9V w 2.5-3-2-6

I'm gonna try to see what I can do at 2.7.

Do you have any tips on which latencies are the most important to keep low, and which ones I can play with more?

Also, I did do my research, and found many people say that the NF7-S was able to get great overclocks and was very stable with many BIOS options for revision 2, so that's what I got. However, I wasn't able to find any solid reviews of this RAM on the NF7-S, and I thought I remember reading reviews that were able to get a high FSB with other RAM, so I figured this RAM could do it.

I suppose I could have done more research, but with so many reviews saying so much about this RAM, I never doubted that the NF2 would have such a negative difference.

One more question: Where do you guys learn all this crazy stuff about TCCD and memory. What is TCCD anyway, all I can find out about it is that it's a type of Samsung RAM.

lllsynthtaxlll
 
lllsynthtaxlll said:
I'm not sure where you got the impression that I was able to POST at 240, I wasn't.

So far the top I can get is 225 @ 2.9V w 2.5-3-2-6

I'm gonna try to see what I can do at 2.7.

Do you have any tips on which latencies are the most important to keep low, and which ones I can play with more?

Also, I did do my research, and found many people say that the NF7-S was able to get great overclocks and was very stable with many BIOS options for revision 2, so that's what I got. However, I wasn't able to find any solid reviews of this RAM on the NF7-S, and I thought I remember reading reviews that were able to get a high FSB with other RAM, so I figured this RAM could do it.

I suppose I could have done more research, but with so many reviews saying so much about this RAM, I never doubted that the NF2 would have such a negative difference.

One more question: Where do you guys learn all this crazy stuff about TCCD and memory. What is TCCD anyway, all I can find out about it is that it's a type of Samsung RAM.

lllsynthtaxlll

tccd is just the chip, its like honda (samsung) and a s2000 (tccd). I beleive its .11 micron (correct me if Im wrong someone?). Also, try running a modded bios. From what I have seen to hit high fsb on nf7 you need a modded bios. Tictac makes good bioses. AHHH reread your post, you couldnt post at 240. Sorry was in a hurry ;) Yea, you prolly need a modded bios.
 
So I tried 230 at 2.7, 2.8 and 2.9 with all sorts of different timings, and was able to get into windows ONCE, but then it crashed. It does seem that 2.7 is more stable than higher voltages at certain timings, and sometimes looser timings seem less stable than the tighter ones, so basically I keep getting random results, but the commonality is that I can't get into windows with 230.

So I guess I'm just boned, and have to accept the fact that I took a risk, and lost. What's surprising is that I'm able to get the same results as everyone else at lower frequencies, but can't even get results at higher ones.

I'm also confused about the fact that I heard that these sticks are really like DDR500 sticks that were relabeled and work at really low latencies at a lower frequency. So if that's true, why won't they work at DDR500? It they could have been sold as such, they would have to have worked on any platform at that speed.

Also, many reviewers basically guaranteed DDR500 speeds, and no one mentioned that it might not be possible on an AXP system.

So although it seemed like I took a small risk, I ended up getting bent over.

Live and learn.

lllsynthtaxlll
 
Most all the reviews and even from the company that assembled this ram module like ocz based their testing on Intel 875 or nforce3 chipsets. At least you have ram chips that gonna show its power when you upgrade to higher specs system. It reminds me when I bought Corsair xms3500 c2 and ran it with xp and nforce2 mobo and got really bad oc with 2.7V. Turn out that sucker is winbond BH5 chip that oc like a beast with +3.0V on Nforce3 or Intel 875 :D
 
Thanks,

I will try and take solace in that. How did you know that they test mostly on those chipsets?

lllsynthtaxlll
 
honestly, I am sick of people thinking that if they see if on a review they can get it. First, you got past the stock rrated speeds....yes...so it did what it was sold to do...granted ocz you are expecting more....so let say ddr 450mhz or 225 a good oc...you got past that....I can say for certain having had 5 abit nf7-s rev 2.0 and none went past 460 without modding it is your board holding you back...and if for some miracle...it is the chip...do you have thd L12 mod done on it....

COMON>...it isnt OCZ's fault you cant get it higher....you should be happy you got it that high...you are not running a intel boad and chip...that is great for a amd non 64...
 
c00z,

I very carefully explained that I'm upset that it's the motherboard/AXP system that's holding me back. I understand it's not OCZ, and I was very sure it was not my sticks. I also explained why I was fairly sure that I would get good results, even though I weren't using an Intel board.

For what I'm getting out of these sticks, I could have gotten cheaper RAM that would do the same - that's why I'm upset, I spent the extra money not because of the low timings, but because it would overclock well, that's what it's marketed for, and that's what all the reviews said it would do. I'm not "happy" that it does stock speeds, that's not what I spent the extra money on.

I understand where I went wrong, and next time I will do more research, but after reading the reviews, which mainly said this RAM is absolutely amazing, who was I to disbelieve them? It would have been nice in one of those 12+ reviews to say a little something about how not to expect the same out of an Athlon XP.

"honestly, I am sick of people thinking that if they see if on a review they can get it"

Then what's the point of a review, let alone the over a dozen reviews on this exact memory. I'm sorry if I somehow insulted you, but I said it before that the blame lies on my shoulders, so thanks for seeing fit to insult me.

centvalny,

Read what? I know the reviews test mainly on those chipsets (I did read those), I meant how do you know that OCZ/Manufacturers test mostly on those chipsets. The OCZ website said Athlon XP/64 and Intel alike.
 
Have you tried 3-4-4-8 or 3-4-4-10? Still won't go near 240? Just as a long shot, what do you have the multiplier at?
 
lllsynthtaxlll said:
c00z,

I very carefully explained that I'm upset that it's the motherboard/AXP system that's holding me back. I understand it's not OCZ, and I was very sure it was not my sticks. I also explained why I was fairly sure that I would get good results, even though I weren't using an Intel board.

For what I'm getting out of these sticks, I could have gotten cheaper RAM that would do the same - that's why I'm upset, I spent the extra money not because of the low timings, but because it would overclock well, that's what it's marketed for, and that's what all the reviews said it would do. I'm not "happy" that it does stock speeds, that's not what I spent the extra money on.

I understand where I went wrong, and next time I will do more research, but after reading the reviews, which mainly said this RAM is absolutely amazing, who was I to disbelieve them? It would have been nice in one of those 12+ reviews to say a little something about how not to expect the same out of an Athlon XP.

"honestly, I am sick of people thinking that if they see if on a review they can get it"

Then what's the point of a review, let alone the over a dozen reviews on this exact memory. I'm sorry if I somehow insulted you, but I said it before that the blame lies on my shoulders, so thanks for seeing fit to insult me.

centvalny,

Read what? I know the reviews test mainly on those chipsets (I did read those), I meant how do you know that OCZ/Manufacturers test mostly on those chipsets. The OCZ website said Athlon XP/64 and Intel alike.
Call them and ask for the solution or....rma it. You can set your goal of oc based on your other component capabilities. I set my goal high b/cause I know my components can deliver and I test it on nforce2 system too.
 
lllsynthtaxlll said:
Well, thanks for all your help, but I'm just p*ssed that I spent the extra $$ on this RAM, and was basically mislead. I knew that there was a chance since all the reviews were on Intel that I wouldn't get as good an overclock, but not like this, this is ridiculous. I could have saved a lot of money getting some different sticks that would have performed the same for me.

And if I have 432, then it really p*sses me off when companies release a product that's great, then after it's been reviewed, puts in inferior parts.

I might spend some more time tweaking, but in the end I'm not getting anything close to what I expected, and would warn against buying this RAM if you have an Athlon XP system.

lllsynthtaxlll
Greetings. I have a lot of things to say, but first of all... you have a few misconceptions. Memory frequency is not about the MAX FSB your motherboard can obtain. for Instance... if you were to think that the INTEL reviews are running 275 1:1 you're almost certain to be wrong (unless the reviewer is using a lower chip such as the 2.4c with a low multiplier)... if you were to have a 3.0ghz chip for instance run a 275 fsb you'd have
3000/200 (to find the multiplier) = 15
15 * 275 = 4125. If you think a 4125 is easily obtainable... you've got anothe think coming. The reason these reviewers can hit such a high FSB is because (in general) people with super high memory frequencies (in general i'm talking about Intel setups, since AMD systems do not like running at a divider) are running with a divider, or a "ratio". This means for every cycle the "FSB" goes through, the memory frequency can be run a-synchronously from the FSB. You have many different ratios such as 5:6, 1:1, 3:4 etc.

For every 'x' FSB cycles you can run the memory 'x' cycles.
for instance... if you were to use PC3200 (200mhz) RAM, if you were to run 1:1
you'd run
FSB = 200
RAM = 200
if you were to run 5:6 ratio
You'd run the:
FSB = 200
RAM = 240
Etc. I hope you get the idea by now. First of all. your MOTHERBOARD is holding you back, as AMD XP systems in general do not run super high FSBs.
The NF7-S is known as a GREAT overclocking board, do not feel cheated of your purchase. It is one of the more solid, and fully featured boards of the NF2 chipsets. I've owned a DFI Infinity, (and still do)... The Infinity is capable of hitting 250-260 fsb right out of the box, and has higher voltage options than most other motherboards i've seen, but are plagued with minor issues (research if you want to knwo about them). The DFI's break-neck frequency features i feel are equal to the NF7-S's total bundle. IMO, both boards have their caveats, and if i were to do it all again, i'd go with the NF7-S.
The general results from a stock "out of the box" NF7-s are about 230fsb.
Some lucky/more 'knowledgeable' users are able to hit 240 or so...
240 is almost easily attainable with a volt mod of some sort (or more) to your board. From what i gather, you haven't volt modded before... but it's fine.
If you haven't played with voltage options in your BIOS it is probably time to do so.
Samsung chips in my experience dont' really respond all that well to voltage and scaling (frequencies). Most of the Samsung chips i've worked with operate best at around a 2.8v, but i haven't worked with their newer TCCD chips, so i won't claim to know about them...

For an AMD AXP system, you're better off running LOW latencies and a 230 or so FSB, in my honest opinion, than running a 275 or so memory frequency.
To be honest, 'recommending' things like "this ram is not for AXP systems" is very arrogant. The reason Samsung TCCD chips are so popular nowadays is because they can be run very well for both Intel, AND AMD (AXP) systems. AXP systems benefit (more than Intel does) from tight timings, and a high-as-possible FSB doing so. A 230 fsb or so 1:1 with 2-2-2-11 timings (11 just works better with NF2 chipsets, dont ask me why) will net you better results IMO than a 250 3-3-3-8 (at the same processor frequency).
Intel systems benefit from raw memory bandwidth. über frequencies on the memory, and running at a divider doesn't take away performance half as much (if at all) compared to an AMD system.
I just thought i'd clear that up, because 'uneducated' remarks like that seriously damage a companies reputation... (dont take it the wrong way, either... i don't mean to call you uneducated/insult you, but obviously there are things you dont know about ram and cross platform performance...)

lllsynthtaxlll said:
c00z,

I very carefully explained that I'm upset that it's the motherboard/AXP system that's holding me back. I understand it's not OCZ, and I was very sure it was not my sticks. I also explained why I was fairly sure that I would get good results, even though I weren't using an Intel board.

For what I'm getting out of these sticks, I could have gotten cheaper RAM that would do the same - that's why I'm upset, I spent the extra money not because of the low timings, but because it would overclock well, that's what it's marketed for, and that's what all the reviews said it would do. I'm not "happy" that it does stock speeds, that's not what I spent the extra money on.

I understand where I went wrong, and next time I will do more research, but after reading the reviews, which mainly said this RAM is absolutely amazing, who was I to disbelieve them? It would have been nice in one of those 12+ reviews to say a little something about how not to expect the same out of an Athlon XP.

"honestly, I am sick of people thinking that if they see if on a review they can get it"

Then what's the point of a review, let alone the over a dozen reviews on this exact memory. I'm sorry if I somehow insulted you, but I said it before that the blame lies on my shoulders, so thanks for seeing fit to insult me.

centvalny,

Read what? I know the reviews test mainly on those chipsets (I did read those), I meant how do you know that OCZ/Manufacturers test mostly on those chipsets. The OCZ website said Athlon XP/64 and Intel alike.
Another clarification: AMD systems can run at a multiplier TOO. (surprise ;D)
AMD (when i say AMD, i mean AXP systems in general, as i'm not all that knowledgeable on the A64 series) get a decently sized performance hit while running the busses a-synchronously, so i wouldn't recommend it... but you CAN. if you want to verify that the sticks CAN run said frequency, or if you just want to make your purchase feel worthwhile ;)
Try running a 220 Fsb or so with a 5:6 frequency
FSB : 220
MEM : 260 or so

as far as "overclock well, that's what it's marketed for"
overclocking is not all about raw MHz. There are many many more facets to overclocking you need to consider, especially since you have an AXP system. IMO AXP overclocking is more artistic than a cave-man-like painting style Intel O/C.

On an intel system (feel free to dispute this anyone), you're limited to raising the FSB since the multiplier is locked. Raise FSB, pump volts when stability gives issues, set.
(obviously over simplified, and the appreciation of true tweaking is overlooked), but there are MANY MANY more variables to consider with an AXP system. Satisfaction level when stable: HIGH. (IMO).

Your version of overclocking should be more to streamline performance rather than pump out raw MHz/high bus speeds. It's a common misconception, and i made the same mistake early on...

Another thing: you make it sound like AXPs give 'poor' results... i'm just beating a dead horse here, but really... 'absolutely amazing' results on an AXP are different than an Intel system. The fault is not of the platform, but yours. If you buy ram that can run 250 "guaranteed" then stop trippin' if your motherboard can't hit the frequency... really you should know the general results/capabilities about a platform before purchasing... this is kind of "hindsight is 20/20" preaching, but first you complain about spending too much money on high quality RAM (which whether or not you know it yet, is doing EXACTLY what you want ;)), but then you complain about owning a platform that can't utilize it to its max FSB.
It sounds like i'm chewing you out, but i dont mean it that way... If anything, i'd like you to read this post and understand that there are different approaches to overclocking...

"Read what? I know the reviews test mainly on those chipsets (I did read those), I meant how do you know that OCZ/Manufacturers test mostly on those chipsets. The OCZ website said Athlon XP/64 and Intel alike."
Just to reiterate: Your board can probably run the memory at 250... it'd just be stupid to do so...

I've been ranting a little too much, and it is getting late... i hope this makes more sense to you when you read it...
Good luck tweaking!
 
I dont kn ow if you ever noticed but the reviews were all doen with p4s and athlon64s, which are able to achieve high fsb's. I dont think i've every heard of a xp going over 250mhz. that would be fucking crazy
 
I've got 4x512 MB of the TCCD stuff running on my ASUS P4C800-E at 250 FSB for months now, it's defying the laws of physics with gusto. I had bought 4x512 MB of OCZ PC-4400 copper because I had needed more than 1 GB and I still wanted to keep my overclock, but it wouldn't boot a fraction past stock. Turns out (at least with 875 boards) that they can't supply the voltage to more than 1 GB, either in 2x 1 GB or 4x 512 MB sets. Picked up the TCCD and it ramped right up past 250, all 2 GB, with nice timings and decent voltage, too. Lost 80 bucks on the restock fee for 4x512 MB of PC-4400 to learn that lesson ; )

Did you try running it at 5:4 or 4:3 and seeing if it was the processor that was limiting the OC? If it was, you might find a (possibly lower-clocked) chip that can OC higher, getting you your 250 (which would probably improve performance overall so you wouldn't notice the difference between your chip running at your max stable and the new chip running higher). If this has been suggested already, I apologize, it's late (or is it early?) for me and I'm only half-reading the replies.

Anyway, good luck on your endeavor, I never tried overclocking my Athlon-32's but I'm sure it's a load and a half of fun to try ; )

If you never do reach your goal, you might consider upgrading to a year-old MB and CPU, like the Intel P4-C's if you aren't an AMDroid. They're famous for their complete lack of regard for the speed rating on the box. You could probably even find one in the for-sale section here since so many people are switching to A-64's now, bless their filthy-rich little hearts ; )
 
diredesire,

I know about dividers, Intel and AMD, I know about Volt modding and using modified BIOSes; I've been overclocking for years and nothing you said to me in that long post was news. 230 at 2-2-2-11 is completely out of the questions for my system. Most if not all the reviews I read used a 1:1 timing, because they are enthusiast sites, and wanted the best performance. Read some of them here:

http://www.ocztechnology.com/products/memory/ocz_el_ddr_pc_3200_platinum_revision_2

I do take offense to being called uneducated, when my simple mistake was not researching enough the limitations of my system - although I never expected to get higher than 250 anyway, knowing that would be amazing, I did expect to be able to get a stable 225 - which isn't too far fetched. Anyway, I've fully admited my fault SEVERAL FREAKIN' TIMES, and am getting tired of it.

I understand that I should have tried to research more on how far AXP systems can be pushed, but I had good reson to expect I'd be able to get at least 230, which I can't even get stable, and as it turns out, 225 gives me an error in Prime95 no matter what voltage/timings I use. Many people here were even surprised that I couldn't get 240.

When I say I do not recommend this RAM for an AXP system, I mean it - don't accuse me of hurting their reputation (like little old me has that power anyway). For Intel and A64 systems, this memory is just amazing, but because I hit a wall so soon, and have to loosen the timings anyway to get above 215, I could have gotten cheaper RAM that would do the same at those speeds. That's why I don't recommend it, not beacuse it's bad RAM, but because for what you get out of it, it's not worth the extra cost compared to some other sticks on my system.

Anyway, thank you all for your help. I've certainly learned a lot more about the limitations of AXP systems, and when I have more time I will tweak my heart out and hopefully get a little further.

lllsynthtaxlll
 
Dephcon said:
I dont kn ow if you ever noticed but the reviews were all doen with p4s and athlon64s, which are able to achieve high fsb's. I dont think i've every heard of a xp going over 250mhz. that would be fucking crazy

I have seen athlon xp's stable up to 270 (infinity and bh5 and ocz ddr booster)
 
My system was :
Abit NF7-S 2.0
XP2600 mobile
1024 OCZ EL DDR400 in dual channel

Max OC was 2.66G (213 x12.5) with ram timings of 2.5-3-3-11 2.8V. The Nforce2 seem to like 11 on the timings. I could have tried for a lower multi and higher FSB but switched to A64 berfore getting around to it. Not even sure if I needed 2.8V on the ram.

The ram was always super stable in mem test 5. It def could have kept on going up.

Still using that ram in my current A64 but havent started OCing yet so no word on its performance
 
lllsynthtaxlll said:
diredesire,

I know about dividers, Intel and AMD, I know about Volt modding and using modified BIOSes; I've been overclocking for years and nothing you said to me in that long post was news. 230 at 2-2-2-11 is completely out of the questions for my system. Most if not all the reviews I read used a 1:1 timing, because they are enthusiast sites, and wanted the best performance. Read some of them here:

http://www.ocztechnology.com/products/memory/ocz_el_ddr_pc_3200_platinum_revision_2

I do take offense to being called uneducated, when my simple mistake was not researching enough the limitations of my system - although I never expected to get higher than 250 anyway, knowing that would be amazing, I did expect to be able to get a stable 225 - which isn't too far fetched. Anyway, I've fully admited my fault SEVERAL FREAKIN' TIMES, and am getting tired of it.

I understand that I should have tried to research more on how far AXP systems can be pushed, but I had good reson to expect I'd be able to get at least 230, which I can't even get stable, and as it turns out, 225 gives me an error in Prime95 no matter what voltage/timings I use. Many people here were even surprised that I couldn't get 240.

When I say I do not recommend this RAM for an AXP system, I mean it - don't accuse me of hurting their reputation (like little old me has that power anyway). For Intel and A64 systems, this memory is just amazing, but because I hit a wall so soon, and have to loosen the timings anyway to get above 215, I could have gotten cheaper RAM that would do the same at those speeds. That's why I don't recommend it, not beacuse it's bad RAM, but because for what you get out of it, it's not worth the extra cost compared to some other sticks on my system.

Anyway, thank you all for your help. I've certainly learned a lot more about the limitations of AXP systems, and when I have more time I will tweak my heart out and hopefully get a little further.

lllsynthtaxlll


due I have told you this about 4x, you need a modded bios to get high fsb out of a nf7. Tictacs modded bios's use tables from dfi's infinity bios. Flash the bios to a tictac and you should be able to get at least 235, also do the l12 mod. You keep on saying it is not good ram for axp system but I see ppl using tccd and hitting 250 on axp system.
 
Back
Top