*** The Official [H]ard ORB ***

cornelious0_0 said:
No kidding, I'd be doing the same thing myself at that point...without question.
Click to expand...

If it can pass a LH or Nature, it's stable enough for me. I usually run a SiSandra mem test, and if that passes...it's on to 2001. Knowing which test stress which part of the system really help, because then you know where to be cautious, and where not to.

I always hold my breath during a LH test, because I know if I pass that one I'm home free :D
 
You guys know what I think. If you have to stop between runs, its not stable. I guess thats why Ill never be number one. I'm happy with the top five. :rolleyes:
 
Why have to stop beetween runs? :confused: If the system can finished straight/normal run then it passed ;)
 
centvalny said:
Can't "stop and run" with AM3 :D
Click to expand...
Not only that, but that silly benchmark actually starts heating things up just sitting there idle, before benching even starts! :rolleyes:
 
uclajd said:
Not only that, but that silly benchmark actually starts heating things up just sitting there idle, before benching even starts! :rolleyes:
Click to expand...

That is true....there's different degrees of stability involved in all of this. If someone's on my tail I'm most likely gonna go for a "balls to the wall" approach and pull out all the stops....whereas if I wanna have my comp more or less stable all around I'll make sure it can get through SuperPi 32M or the hour long S&M stresstest (if I'm short on time or impatient).....or one full pass of Memtest and 12 hours of Prime95.

I like to know what my all out speeds are that I can either boot into window at, or run SuperPi 1M.....but beyond that it is still important to make sure you're still taking your time to find your max stable clocks.....well, unless you're using a seperate benching rig that you don't USE 24/7. ;)
 
centvalny said:
Why have to stop beetween runs? :confused: If the system can finished straight/normal run then it passed ;)
Click to expand...

When it comes to my gaming rig, I'm alot harder on stability. When it comes to my benching rig though, it's whatever gets the highest score.

Everyone has their own degree of how far they will go. Some people throw fans on the condensor, open windows, put their comps in refrig units, etc...to get those extra few MHz...If I have a few thousand invested in coolers, chips, cards, etc...why wouldn't I do everything I can to get those last few points? Volt-modding is something we do to get extra scores, yet it seems to be accepted...when stopping between runs isn't?

To each his own I guess, I think sometimes we all need to step back and look at it from another perspective. To some people this is more than a hobby.
 
I won't ever post a benchmark if the system doesn't also pass stringent tests, at least a couple of full memtest passes plus a day or so of Prime95, not to mention the other culprits. I think it's dishonest to publish a benchmark if the system being benched isn't 100% stable. That's just my opinion though.
 
ThomasE66 said:
I think it's dishonest to publish a benchmark if the system being benched isn't 100% stable. That's just my opinion though.
Click to expand...
Oh pul-eeeze. :rolleyes: This is who can post the highest number.

One flaw in these [H] contests is the rules keep changing. We get through 10 pages of posts, and suddenly there are new standards? After someone kicks the ball, you can't move the goal posts.

Where in the first post does it say anything about being 100% stable? :confused:
 
uclajd said:
Oh pul-eeeze. :rolleyes: This is who can post the highest number.

One flaw in these [H] contests is the rules keep changing. We get through 10 pages of posts, and suddenly there are new standards? After someone kicks the ball, you can't move the goal posts.

Where in the first post does it say anything about being 100% stable? :confused:
Click to expand...
Its actually "free for all" contest and theres no standad of operation. Its just my personal preference when runnin benches. Never ran memtest or prime...those are for labs. I want it fast and if it not, then up the vcore and vgpu...set higher htt and go :D Quick and simple.
 
uclajd said:
Oh pul-eeeze. :rolleyes: This is who can post the highest number.

One flaw in these [H] contests is the rules keep changing. We get through 10 pages of posts, and suddenly there are new standards? After someone kicks the ball, you can't move the goal posts.

Where in the first post does it say anything about being 100% stable? :confused:
Click to expand...

Where in my post did I say it was the rule? I was talking about my personal preference. Reading is fundamental ;)

EDIT:

Maybe I was not eloquent enough in my post. I don't care what the rest of you do. Clock your system so high that all you can do is barely finish a benchmark run before it blue screens. *I* feel that for *me* it would be dishonest to represent my machine as performing at a level where it's unstable. That's all. I don't judge what others do. When I'm perfect then maybe I'll start doing that.
 
I just bristle when I see the term "dishonest." If the thread rules in the first post had stated "stable" benchmarks, that would be one thing - not that stability could possibly be verified. But like every other benchmarking contest, the point is to post the highest number.

Of course I don't run bleeding edge 24/7. But this is a contest, the rules have been stated (or not), so let's not try to move the goal post after the ball is kicked. :p

I will say that I don't put unstable CPU/GPU clock numbers in my sig. Some people I have seen actually take the absolute max screenie under clockgen and stick that in their sig. Now that's silly.


ThomasE66 said:
Where in my post did I say it was the rule? I was talking about my personal preference. Reading is fundamental ;)

EDIT:

Maybe I was not eloquent enough in my post. I don't care what the rest of you do. Clock your system so high that all you can do is barely finish a benchmark run before it blue screens. *I* feel that for *me* it would be dishonest to represent my machine as performing at a level where it's unstable. That's all. I don't judge what others do. When I'm perfect then maybe I'll start doing that.
Click to expand...
 
I didnt mean to get everyones "panties in a wad" with my post. I do it my way, and you do it yours. I am trying to make the most 24/7 stable rig I can. Thats what I want to do. I didnt mean to diminish anybodys scores or benching methods. Like I said earlier, I am happy where I rank. Maybe someday I might start "caseless" benching. But for now I prefer to bench "in the box" rather than out of it. :D SIK
 
LOL, James you better get a bigger case so you can use the chiller I'm putting together for you :D Hmmm...chiller on top, Vapo on bottom?

PM me here or at VO when you get a chance, I've got some plans.
 
VO_Nuclear said:
LOL, James you better get a bigger case so you can use the chiller I'm putting together for you :D Hmmm...chiller on top, Vapo on bottom?

PM me here or at VO when you get a chance, I've got some plans.
Click to expand...

Wow.....sure SOUNDS nice.....can't wait to see it. :D
 
Don't expect to see anything soon. Making a compact yet effective GPU chiller for James is going to be a chore. I went a picked up some compressors tonight, but they are all too big for what I'm wanting to do. Once I get the compressor picked out, I'll start a thread in the Extreme cooling section.

On a side note...I'm getting really pissed & impatient at PC Power. They have had my PSU for 1 1/2 weeks and still don't have it ready. It was supposed to be here friday, then I got a call saying today...now they are saying anywhere from Thursday to next Tuesday. It sucks being held up by a PSU, I could've bought a new one by now :(
 
There will be a lot of changes coming after this weekend. The "new card" will be released, and it will meet/beat our efforts today, stock. At least we know the reason for cancellations of pre-orders for the x850's along with the very limited release of the card into the retail channel. I think ATI has been producing the cards, packaging them and holding them, using up production capacity. No paper release, this time.

BTW, the SLI cards are gone at this moment. I am running an awesome Matrox Millenia PCI card, right now. ;)

And, to counter the stability comment earlier, if it passes through the bench, that's good enough for me. Besides, Prime fails almost immediately on my system yet I can run any game for hours on end at my 24/7 setting. That's stable enough for me.

I am not in the "bench rig only" club. My system is used for all my activities and is actually an encased PC. But, that will reduce me to also ran status, which is fine with me. All the big boys have the "naked" PCs for running benchmarks. Makes sense, since it offers easy component access for supercooling. And, they run each component of the benchmark separately then compile the result for a score. They do this on 01 and 03. So, they have a system config for each piece. That's a lot of work.

So , hanging in the lurch to build the next system config. Just have to wait for the right card. So, in pit row, right this moment, watching the cars zip by...
 
HeavyH20 said:
So , hanging in the lurch to build the next system config. Just have to wait for the right card. So, in pit row, right this moment, watching the cars zip by...
Click to expand...

That's exactly what I'm doing....even though I should be getting some nice results out of my 3200+ in the Ultra-D with my PC4800 GSkill, I'm gonna stick in the ABit uGuru X300 and go for some X300 world records while I'm waiting to pick up the r520. :cool:
 
HeavyH20 said:
There will be a lot of changes coming after this weekend. The "new card" will be released, and it will meet/beat our efforts today, stock.
Click to expand...
Sorry, I haven't been following the game - what are you referring to, a revision or a new GPU?

And, to counter the stability comment earlier, if it passes through the bench, that's good enough for me. Besides, Prime fails almost immediately on my system yet I can run any game for hours on end at my 24/7 setting. That's stable enough for me.
Click to expand...
Concur in part, but I have found that video encoding is the true torture test, IMO. The bandwidth of crushing a DVD file seems to make the difference. To me, it's analogous to amperage, the amount of "stuff" flowing through thye CPU really heats things up. How could mere number crunching be as bad, even if you are using SSE patches?

Speaking of which, this week I encoded Collateral @ 80% in 7 minutes on this rig. I'd like to an AMD do that that isn't on LN2. :D

I am not in the "bench rig only" club. My system is used for all my activities, and is actually an encased PC.
Click to expand...
Me too. I do work, gaming, encoding, etc on this thing. I distinguish between the dry ice and LN2 gimmicks vs. real systems that sit on people's desks.

172W pelt and 24V Mean Well arrived today. I think we might have some new scores coming soon. ;)
 
cornelious0_0 said:
Haha....gonna be kinda tough when i dont have the card anymore....keep with the times man. ;)
Click to expand...

You beat me to it, I figured that out about a minute ago. Give me a break; I've had a few beers! ;)
 
Fun fun fun :)

I took almost a year off from major tweaking and overclocking, and probably longer than that from this place. Hence the new name and userid, the old one got purged last year. I couldn't stay away forever though. Crisis after crisis at work and with my family is to blame. It's fun to be back.

Ok, sorry to get off topic.
 
ATI will be giving a peak at a new video card this weekend that has 512 MB of RAM and an unkown core. It may be just an x850 with more RAM, or it could be a new core. Makes sense, since why spend all the extra money on a new fab for the existing release of x850's. They were just a warmed over version of the existing core. The hint is that they seem to clock to 650 on the core rather easily with the stock HSF, the rumored spec of 700 MHz on the new core does not sound that unreasonable. May even be a 24 pipe card. Either way, ATI is looking to set new WR's on 03 and 05 with a single card. And, that is considering the SLI systems. Should be interesting to see what they release.
 
HeavyH20 said:
ATI will be giving a peak at a new video card this weekend that has 512 MB of RAM and an unkown core. It may be just an x850 with more RAM, or it could be a new core. Makes sense, since why spend all the extra money on a new fab for the existing release of x850's. They were just a warmed over version of the existing core. The hint is that they seem to clock to 650 on the core rather easily with the stock HSF, the rumored spec of 700 MHz on the new core does not sound that unreasonable. May even be a 24 pipe card. Either way, ATI is looking to set new WR's on 03 and 05 with a single card. And, that is considering the SLI systems. Should be interesting to see what they release.
Click to expand...

Being hardware agnostic, I love the ATI-NVidia competition as much as I love the Intel-AMD competition. It pushes prices down and performance up for one and all. I have ATI and NVidia cards in the house, and I'm watching carefully for future developments for my next upgrade :)
 
gclg2000 said:
I have a feeling its going to be more than a 512mb x850
Click to expand...
That's all that is being reported by the rumor sites (850 w/ 512), and of course they are never wrong. :rolleyes:
 
You mean a site like these below ??

http://www.theinquirer.net/?article=20896
or
http://www.theregister.co.uk/2004/12/21/nvidia_nv48/ and http://www.theregister.co.uk/2004/11/03/ati_tapes_out_r5xx/
or
http://www.megagames.com/news/html/hardware/atir520withshader30.shtml
or
http://bjorn3d.com/theforum/viewtopic.php?t=59134

(all rumors of course....)

24 "Pipelines"
32 Texture Units
96 Arithmetic Logic Units (ALU)
192 Shader Operations per Cycle
700MHz Core
134.4 Billion Shader Operations per Second (at 700MHz)
256-bit 512MB 1.8GHz GDDR3 Memory
57.6 GB/sec Bandwidth (at 1.8GHz)
300-350 Million Transistors
90nm Manufacturing
Shader Model 3.0
ATI HyperMemory
ATI Multi Rendering Technology (AMR)
Launch: Q2 2005
Performance: Over 3x Radeon X800 XT !!! (for single R520)
16x stochastic FSAA
FP32 blending, texturing
Programmable Primitive Processor/Tesselator
 
gclg2000 said:
I have a feeling its going to be more than a 512mb x850
Click to expand...

Sorry guys, it's NOT an X850 /w 512MB of ram....Zardon doing the testing openly and publically stated that it has NOTHING to do with current hardware with more memory.
 
I'm well aware of the 520, but these sites all say Q2 2005 (Mayish). This is Q1. :confused: :confused:

Considering the last two major ATI launches have essentially been paper (800 XTPE and 850), I'm a little skeptical that this thing ships in numbers Q1, regardless of what a few select gurus happen to be testing. :p

Hope I'm wrong - I'll run out and buy one. ;)

JD

gclg2000 said:
You mean a site like these below ??

http://www.theinquirer.net/?article=20896
or
http://www.theregister.co.uk/2004/12/21/nvidia_nv48/ and http://www.theregister.co.uk/2004/11/03/ati_tapes_out_r5xx/
or
http://www.megagames.com/news/html/hardware/atir520withshader30.shtml
or
http://bjorn3d.com/theforum/viewtopic.php?t=59134

(all rumors of course....)

24 "Pipelines"
32 Texture Units
96 Arithmetic Logic Units (ALU)
192 Shader Operations per Cycle
700MHz Core
134.4 Billion Shader Operations per Second (at 700MHz)
256-bit 512MB 1.8GHz GDDR3 Memory
57.6 GB/sec Bandwidth (at 1.8GHz)
300-350 Million Transistors
90nm Manufacturing
Shader Model 3.0
ATI HyperMemory
ATI Multi Rendering Technology (AMR)
Launch: Q2 2005
Performance: Over 3x Radeon X800 XT !!! (for single R520)
16x stochastic FSAA
FP32 blending, texturing
Programmable Primitive Processor/Tesselator
Click to expand...
 
uclajd said:
I'm well aware of the 520, but these sites all say Q2 2005 (Mayish). This is Q1. :confused: :confused:

Considering the last two major ATI launches have essentially been paper (800 XTPE and 850), I'm a little skeptical that this thing ships in numbers Q1, regardless of what a few select gurus happen to be testing. :p

Hope I'm wrong - I'll run out and buy one. ;)

JD
Click to expand...

It was my understanding that they are just giving a peek at an upcoming card, not announcing the availability of one.
 
May is 2 months away.....seems like a great launch exhibit to show off their new card 2 months before its released at their first "OC'ing competition" And possibly give some more time to fix bugs.....but let's be honest, how hard was it at first to get an xt pe when they came out, same shit gonna happen
 
You must log in or register to reply here.
Back
Top