PRIME Don't Place Much Faith In It?

[Guest DasFox]

Someone told me this:

 

That's a CPU error IRRC. If you get a RAM error it gives you a much longer message.

 

But really, prime stability means nothing to me. If it can do what I want without crashing, I don't care if it fails instantly. So don't place too much value on prime stability, test it by doing what you'd actually want the thing to do.

 

My reply:

 

I thought people SWEAR by PRIME, that if you get any errors then there is a miscalculation on the hardware, and it's not running correct.

 

See being somewhat of a PRIME newbie in understanding it this way I don't really know, but I'm not a newbie as far as the program is concerned, I've known it for years.

 

Well the thing is I ran PRIME on my box one time without any overclock, and with it all stock, and from what I can tell if someone has a stock system then PRIME isn't going to give errors, and if it does that means you have bad hardware.

 

I don't want to ruin my CPU if it's giving me errors. I thought the hardcore overclocking geeks swore by PRIME, and if you get an error, then it's not correct, and you fix it till it doesn't give you an error.

 

Anyhow as far as killing my CPU is concerned the volts are at 1.5 and it runs cool, 43c idle and 50c loaded with PRIME. So maybe as far as PRIME is concerned even though it gives an error if I can game and never crash and the box seems to run stable, and I have the CPU cool enough maybe it's ok?

 

THANKS

[RubberDuck]

That's a CPU error IRRC. If you get a RAM error it gives you a much longer message.

 

But really, prime stability means nothing to me. If it can do what I want without crashing, I don't care if it fails instantly. So don't place too much value on prime stability, test it by doing what you'd actually want the thing to do.

 

THANKS

 

IMOP that is pure bull .. Your system is not stable if it is not prime stable.

[Guest r3d c0m3t]

Agreed. Why even run Prime95 if you're going to disregard the results. That's not only wasting your time, but the effort put into the program for people like us who actually care whether or not we have a 100% stable OC, or just a stable PC period.

[red930]

First off, I believe the error for a ram or cpu problem is the same within Prime. It has been a while since I've used it however (I use Orthos exclusively now) so I could be wrong. However, a ram problem would likely stop a Prime Blend test much quicker, whereas it'd more than likely be a cpu issue if it fails a few hours into the test.

 

I thought the hardcore overclocking geeks swore by PRIME, and if you get an error, then it's not correct, and you fix it till it doesn't give you an error.

 

Very much depends on what you mean by 'hardcore'. Many of the top benchmarkers generally don't use it (historically anyways, always exceptions of course) because obviously it'll hinder the top numbers they can get. However, DFI-Street and many of us that populate it (as well as a considerable number of proponents at XS and other larger sites) have pushed for Prime/Orthos stability being more important, as it ensures stability regardless of what you're doing, and more than likely is much, much more demanding than whatever apps you normally run, be it BF2, CAD, Photoshop, etc.

 

There are still some people that refuse to use it, claiming that if it runs games fine then why bother? Well my simple answer is: it's faster. I don't want to have to play a 4 hour run of a video game just to find out it isn't stable. Running even an hour of Orthos or Prime 'blend' is going to be much more strenuous. Obviously even the 8 hours that we recommend for the OCDB, and 12+ hours that others recommend doesn't absolutely GUARANTEE no errors....just makes the chances of it infinitesimally small.

 

As if its even a big deal - everybody sleeps. Let your computer run during that time and you'll have what, probably at least 6 hours completed (assuming its stable).

 

The other issue I take with people not using these programs is just the e-penis of it. Take a C2D, lets say the max I can get completely stable is 3.5GHz, then compare it to the max I can get 'game stable' of maybe 3.75GHz (unlikely, but okay). The difference in speed for the vast majority or people that only game is NEVER going to affect your gameplay. Even a 2.8GHz Opteron vs. 3.0GHz Opteron won't show hardly any improvement at all. So stressing your processor that much is just stupid. Every time an error happens it's just one more chance of an OS installation to go wrong, and the (unlikely, but possible) chance of hardware failure.

 

For the few people that actually do run intensive programs that could actually utilize the extra speed:

 

1. It won't shorten the time of rendering (or whatever calculation you're doing) all THAT much

2. Even a more likely chance of something going wrong, what with the extra strain

 

Anyways. Here we always say Prime/Orthos stability is king, and I believe it. I want to have a computer that is usable, and that I don't have to worry about. For those that disagree, that your computer works fine without running these programs....great. Just don't come complain later if something goes wrong.

[My Ugly PC]

Fortunately, how you run your rig is your own business. If you expect to post as an example of good timings or OC results you must be prime stable, especially if your are giving someone advice. But if your results and performance are for your benefit only, there is no discussion, do as you see fit. If you end up with a corrupted OS or program install, then you know you were beyond your limits. I really think most folks would rather just Prime up front and not have to worry about stability. Regarding DFI in specific, a desire to avoid prime may be outright laziness and not wanting to learn all the memory options etc, IMO those people should buy a damn ASUS.

 

Personally, I've had several configurations that I have not bothered to get beyond 2hrs prime stable, for lack of time or whatever (meaning I ended stress tests at the 2hr mark, usually I go for about 4-6hrs, but I will take a really attractive setting if it goes for less). I'm a big boy and I can handle the consequences of living to my own standards. If your not experienced or confident enough to blaze your own path then IMO just opt for safety and prime at least 4hrs, I bet your FPS wouldn't suffer at all. You can easily answer your own questions by 3d benching a "game stable" setting, then dropping to a stable range and repeating your tests.

 

Eh, try to isolate good advice from that of sheeple (ie followers, listeners not experimenters) but don't let anyone tell you how to run your rig. If your happy with some other method don't let someone come in here and wag their finger at you, I don't care who they are. Your rig, your risk and grey area exists in every issue.

 

Ugly

[red930]

If your happy with some other method don't let someone come in here and wag their finger at you, I don't care who they are.

 

Um, actually I wouldn't really recommend doing that here.

 

Don't get me wrong, honestly I don't care how other people overclock, it's their problem and not mine, and certainly your method is quite a lot better than what more than a few people say is stable.

 

However, coming here and claiming an overclock is stable with no proof besides a CPU-Z screenshot or 1M SuperPi run will NOT fly or 4+ hours of gameplay won't fly here. The OCDB is set up for a reason, and especially for those new to overclocking this is our guidelines if you want to post about it. Again, not pointing you out, but there have been others that have come here trying to do such a thing, and their time here is very short-lived especially when only providing CPU-Z screenshots.

 

Keep in mind that this is because this is a support forum, primarily here to help users with DFI products, then as a secondary objective to offer help with overclocking said DFI products. Allowing people to post such ridiculous claims for those new to overclocking just is not wise, as people could seriously damage their hardware without knowing what they are doing.

 

For the 'pros' out there....well good luck.

[Branjo]

If testing an OC then I'll do a quick 30 min OCCT if that finishes and the temps are good then thats enough for me. I'm not very anal about stability.

[My Ugly PC]

Don't worry about quoting me with my name included, I'll own all my words wrong or right. You address a point I thought I laid out pretty clearly and very early in my post (though my speechifying skills have always been suspect, I swear to god in person I sound like a stuttering George Bush impression).

 

Second and third sentence of my OP unedited except for appending [/b] for boldness..

If you expect to post as an example of good timings or OC results you must be prime stable, especially if your are giving someone advice. But if your results and performance are for your benefit only, there is no discussion, do as you see fit.

 

My point is completely about personal use only, not posting any misleading results from it. I didn't think I had to be any more clear than i already was (highlighted in bold). In fact I think there should be bold and clear text in the rules stating that every post in this forum by default is in a support capacity as that is what people come here for and you cannot control who reads your posts, and posting timings or OC results that are not Prime stable is antithetical to this forum's support nature, and help no one. I mean if that is what every mod and admin believes and enforces why isn't it said clearly where it counts?

 

Game and bench and operate how you chose, post only in a responsible way ie only results and timings proven stable that can help someone, thats my position. A black and white prime for 8 hours or your rig will die attitude.. I just don't understand it, overvolt .05v and your risking your hardware, not priming to a long extent and you really risk alot less.

 

I suppose that is the answer to the question that was actually asked, the most likely scenario is that all you are risking is having to reinstall an app or your entire OS. Applying the extra voltage that it takes to get a rig from game stable to prime stable is in all likelyhood much more of a risk to hardware and investment, someone will jump on me if I'm technically wrong about that. Go back to my first post to actually find my personal belief about stability testing, but no, prime stability isn't the gospel unless posting your results is necessary.

 

I hate to get into a disagreement about philosophy, but when it prevents someone from having fun with their own equipment in their own way IMO truth is more important than babying someone, or deciding what is best for them instead of informing them to make their own decision. I don't get the point of creating a generation of guys that do what the dude with 1000 posts says instead of finding their own way if they so chose. It's certainly not an opinion worth being vilified over (not referring to myself at all).

 

Dammit, I knew I would prattle on and talk too much, if anyone wants a second opinion just PM me or go outside of this support forum for advice.

 

Ugly

[faheyd]

Don't worry about quoting me with my name included, I'll own all my words wrong or right. You address a point I thought I laid out pretty clearly and very early in my post (though my speechifying skills have always been suspect, I swear to god in person I sound like a stuttering George Bush impression).

 

Second and third sentence of my OP unedited except for appending [/b] for boldness..

 

 

My point is completely about personal use only, not posting any misleading results from it. I didn't think I had to be any more clear than i already was (highlighted in bold). In fact I think there should be bold and clear text in the rules stating that every post in this forum by default is in a support capacity as that is what people come here for and you cannot control who reads your posts, and posting timings or OC results that are not Prime stable is antithetical to this forum's support nature, and help no one. I mean if that is what every mod and admin believes and enforces why isn't it said clearly where it counts?

 

Game and bench and operate how you chose, post only in a responsible way ie only results and timings proven stable that can help someone, thats my position. A black and white prime for 8 hours or your rig will die attitude.. I just don't understand it, overvolt .05v and your risking your hardware, not priming to a long extent and you really risk alot less.

 

I suppose that is the answer to the question that was actually asked, the most likely scenario is that all you are risking is having to reinstall an app or your entire OS. Applying the extra voltage that it takes to get a rig from game stable to prime stable is in all likelyhood much more of a risk to hardware and investment, someone will jump on me if I'm technically wrong about that. Go back to my first post to actually find my personal belief about stability testing, but no, prime stability isn't the gospel unless posting your results is necessary.

 

I hate to get into a disagreement about philosophy, but when it prevents someone from having fun with their own equipment in their own way IMO truth is more important than babying someone, or deciding what is best for them instead of informing them to make their own decision. I don't get the point of creating a generation of guys that do what the dude with 1000 posts says instead of finding their own way if they so chose. It's certainly not an opinion worth being vilified over (not referring to myself at all).

 

Dammit, I knew I would prattle on and talk too much, if anyone wants a second opinion just PM me or go outside of this support forum for advice.

 

Ugly

 

I gotta tell ya, you're just not getting it and I'm not going to post 6 paragraphs to show you why you are not getting it.

I've had more fun with my PC by following the guides, doing the O/Cing, and listening to the pro's, than trying to blow up my PC on my own. Sure, I like to learn things on my own, but quite frankly , if someone that has already burned up a few boards can tell me the 'right' way to do it, I'm in. No one is telling any user what they 'have' to do here (as far as I can see), they are only showing 'the way', the true path to enlightenment. There are many paths, but only a few will get you to the goal. What's the goal, for me, it is to get the most bang for buck for all the cash I've spent on my rig. If might be something different for someone else. I don't overclock to impress anyone, I just do it because I can and it makes me feel good for some reason. Sorta like tuning up an old car. The guides are just that, guides. They provide a useful bucket of information for the newbs and pro's alike. I don't have enough cash to 'fake' an overclock and blow up a X2 cpu. Haphazardly doing a quicky test because the numbers 'look' good, is a real way to blow your cash down the toilet. Wanna have fun and not pay attention to detail, buy a dell and join AOL.

 

The reason I think this site is where it is today, is because of its strong adherence to standards and not just 'winging' it. Wild butt guesses, unproven results, dangerous practices just don't cut it. Sure, everyone is here for fun, all of us. But we can't have fun if our rig BSODs or melts down.

[My Ugly PC]

@faheyd

 

Whoa, resurrected thread. I took the liberty of rereading it from the top, and I definitely gotta undo the tendencies of persuasive writing that I half learned in school (damn unfinished education). A post isn't an essay, persuasive writing as taught nearly always makes one come of cocky. Stupid Ugly.

 

I have read your post a few times over, and I respect your opinion, and your ability to make it without flaming. And I'm sorry I can't complete my point without multiple paragraphs. But you just made my point for me, you explained it perfectly. That whom I was struggling to explain. The guy who suggests or outright states that not priming a PC for an extended time will damage a computer. How?

 

I just don't understand it, overvolt .05v and your risking your hardware, not priming to a long extent and you really risk alot less... the most likely scenario is that all you are risking is having to reinstall an app or your entire OS. Applying the extra voltage that it takes to get a rig from game stable to prime stable is in all likelyhood much more of a risk to hardware and investment, someone will jump on me if I'm technically wrong about that.

 

I'm no where near as smart as most of the folks here, but that quote is the best my brain can come up with. My mentioning of voltage was merely to prove a point, an example that what we all accept as a very basic part of OC'ing is more dangerous than not priming for 8 hours. Again I want to stress that I'm not one of these guys around here that is an engineering expert, but seriously how is it dangerous to my hardware to prime for 2 hours instead of 8? Software corruption be it app, driver, or OS, yes that is what I risk. It's undoable. Worst case scenario I reinstall OS. So I have to factually technically disagree with you there. Final time I repeat, I hope someone will jump on me if I am technically wrong, neither of us are experts, though we may make the mistake of speaking in absolute terms.

 

I don't have enough cash to 'fake' an overclock and blow up a X2 cpu. Haphazardly doing a quicky test because the numbers 'look' good, is a real way to blow your cash down the toilet. Wanna have fun and not pay attention to detail, buy a dell and join AOL.

 

The reason I think this site is where it is today, is because of its strong adherence to standards and not just 'winging' it. Wild butt guesses, unproven results, dangerous practices just don't cut it. Sure, everyone is here for fun, all of us. But we can't have fun if our rig BSODs or melts down.

 

Moving away from our mere disagreement about the former issue, I regret the above comments and personally feel that none of them relate to me or anything I posted. Seriously, "not just 'winging' it. Wild butt guesses, unproven results, dangerous practices... Wanna have fun and not pay attention to detail, buy a dell and join AOL." I mean come on, except for "unproven results" in certain contexts, what relevance is any of that and how does it relate to my opinion or posts.

 

I get it if you just fell into a pitfall trying to articulate a point, but I maintain that my being different on occasion (you'll find from my first post here that I do believe in priming to an extent, just not a static absolute extent) does not mean I am devoid of skill and method. I know a little bit, jeez :tooth:

 

Ugly

 

 

P.S. Again as this was overlooked last time, I hope prospective respondants understand that I've said all along..

If you expect to post as an example of good timings or OC results you must be prime stable, especially if your are giving someone advice.

[faheyd]

And your post just shows, you don't get it. I'm sorry, I'm not gonna flame you to death or explain it to you. Stop taking it as an insult and try to gain some insight. Not all of it was a finger pointed in your direction. I really do give up and it's not worth hashing over, later.

Dylan

[CharmedLover84]

Okay.

 

I am going to be blunt, then I'll be nice.

 

Bluntly, read the stress.txt

 

It says in PLAIN ENGLISH:

 

"IF YOUR COMPUTER CANNOT RUN PRIME 95 WITHOUT ERRORS AT STOCK SPEEDS YOU DO HAVE A HARDWARE ISSUE"

 

One of your pieces of hardware is faulty, plain and simple.

 

Now, being nice, perhaps you should try testing it in different ways like suggested (Small FFTs for CPU stress, Blend for RAM test) to figure out WHAT SPECIFICALLY is bad in your system.