HardOCP says anandtech has bad methods

[Devo2007]

What I don't like about HardOCP's reviews is simply that I have virtually no idea just what kind of improvement I'd get from upgrading to an HD3870X2 vs. my current 8800GTS 320MB. Sure, they claim to "play through entire games" which would mean it's a lot of work to do what they do (fair enough), but it's absolutely meaningless to me. Not only do I want to know if the 3870X2 is faster than an 8800GTX, but I want to see whether it's worth it to get one of those cards, when an 8800GTS 512MB would be a better value (and I don't want to have to read through three or four reviews on the same site to find that out).

 

[Keysplayr]

Looks like someone is trying to generate traffic for H. Don't fall for it folks.

 

[Zinn]

Here's a good link:
http://www.gpureview.com/hardo...logic-article-630.html

 

[Killrose]

Why HardOCP felt they had to attack AT when there are so many other sites relying on the same benchmark material is bothersome. But I myself do prefere thier reviews over most other reviews on the net including AT. I'm not crazy about their most playable settings comparisons, but they do use actual gameplay, and yes I realise its based on thier version of gameplay. I also like the fact that they also show the FPS graphs and include AA and AF settings which AT does not.

Maybe the review community should get together, come up with a set of in-game/actual gameplay benches to use in addition to the standard benches if they like and use them with a certification of that community. I think this would have been more productive than an attack on AT to come with some sort of review "round table" so to speek and make up a set of "review community approved" benches. However the OEMs would no doubt optomize for these benches, but they would have to write for two sets of benches: the in game bench provided by the game developer and the actual game play bench developed by the community. I think it would be alot harder to optomize for the 2 benches instead of 1.

 

[CP5670]

we already do that ... it is SO easy to make a custom time demo; Keys made a good one for STALKER - the USUAL thing is to POST it for everyone to replicate and for peer review. Kyle's is HIDDEN ... no one can review his benchmarks because we don't know which section of the game he choose.
:roll:

unethical is a word that springs into my mind

... and there is NO way to keep recorded timedeomos from "falling into the 'hands" of the GPU companies ... but it is SILLY for them to optimize for each benchmark - they really need to optimize the game ... and generally they do just that

I don't see how they could get the demos if they aren't published, unless they have a shill among the review staff. It's not ideal to keep them hidden, and the site would need to have some credibility already established (which HardOCP frankly does not), but this would prevent the companies from optimizing specifically for the demos. This situation is still better than having actual gameplay runs that even HardOCP admits can never be the same between different cards.

I also like the fact that they also show the FPS graphs and include AA and AF settings which AT does not.

That is the one thing I do like a lot about their reviews. The graphs are far more useful than a single average framerate score. Are there any sites that benchmark normal timedemos but still show those graphs?

I'm studying mathematics at Helsinki University and for example 3dMark is like most of those things that we are learning them: They don't have any real life value and we'll use them just because someone has invented them.

Nice to see a fellow math student here.

 

[DAPUNISHER]

Originally posted by: keysplayr2003
Looks like someone is trying to generate traffic for H. Don't fall for it folks.

That may indeed be the case, but there is still a legitimate discussion in all of this, don't you agree? You can shoot the messenger, but it serves little purpose. We could all do without Bennett's P.T. Barnum Showmanship ways, but that doesn't invalidate all of his assertions. BTW, Anandtech, having approx. 70% more registered members, is a good guess for why he singled it out, when disparaging the methodology of using a canned bench.


Benchmark specific optimizations suck IMO, and if reviewers/evaluators/whatever they want to call themselves these days, aren't going to do their best to sleuth them out for us, the buying public, they may as well be regarded as corporate shills. My perspective is, do more work on these reviews, and don't pull any punches, ever. Everyone is in such a rush to get their review out, they aren't being as thorough as they could be otherwise.

 

[dennilfloss]

Oooo teh drama...:Q

 

[Keysplayr]

Originally posted by: DAPUNISHER

Originally posted by: keysplayr2003
Looks like someone is trying to generate traffic for H. Don't fall for it folks.

That may indeed be the case, but there is still a legitimate discussion in all of this, don't you agree? You can shoot the messenger, but it serves little purpose. We could all do without Bennett's P.T. Barnum Showmanship ways, but that doesn't invalidate all of his assertions. BTW, Anandtech, having approx. 70% more registered members, is a good guess for why he singled it out, when disparaging the methodology of using a canned bench.


Benchmark specific optimizations suck IMO, and if reviewers/evaluators/whatever they want to call themselves these days, aren't going to do their best to sleuth them out for us, the buying public, they may as well be regarded as corporate shills. My perspective is, do more work on these reviews, and don't pull any punches, ever. Everyone is in such a rush to get their review out, they aren't being as thorough as they could be otherwise.

Absolutely agree. There is definitely a legitimate discussion in this. Just don't run over and register at H just because of this and start anything. That is more or less what I meant by "fall for it".

But you are right.

 

[McCartney]

Originally posted by: apoppin
thanks ... and welcome

i sent a PM to Derek ...


i sense a war brewing ...

i do have to agree with their intro, however:

HardOCP has been thoroughly mired ...

that is the correct term

HARDFUD eh?!

LEGIT BENCHMARKS>

I feel like the guy everyone called crazy, but I'm the genius. Seriuosly apoppin man, why did you have to try to insult my favorite forum? I love AT, but you just had to stab at hardocp. Your beloved review is falsified, and the 3870x2 was never king of the hill. Sorry man.

 

[madh83]

Originally posted by: Rusin
Madh83:
Not entirely on your topic, but this came to my mind:
I'm studying mathematics at Helsinki University and for example 3dMark is like most of those things that we are learning them: They don't have any real life value and we'll use them just because someone has invented them. Also it's like our courses there. They are made from basically two part: First there are lectures (Anandtech) that talk about mathematical theories on and on and talk much about theoretical situations..They don't usually tell how these theories apply in practise. Then there are after these lectures these small group teaching which is all about how these things work in practise (OCP). Thing is that if you test game many times it means that single test doesn't impact on overall score that much that it would make any significant difference.

Well I like Muropaketti's way more.. they try to find setting that runs at 30 fps minimum on that card which is being reviewed and then test other cards with that same setting. Some time the continue testing with the better card to find that "30 fps" for them.

Okay, I'm not completely clear on what you're getting at, and I'm just adding or elaborating on what I said before in a non-confrontational way. What you're assuming with multiple tests is essentially the law of large numbers, I think that's what your getting at without mentioning the theory directly. The thing is, that just tells us an average which may not be descriptive enough, I think adding descriptive statistics for the entire distribution of frame rates would be more helpful. H' does take a step in that direction with their graphs, which I like.

However, I think you were using the law of large numbers to justify that if H had repeated his test over and over it would give us a result with statistical value. My understanding is that H's tests are done by him running around a game with a log of frame rates. This is the part that is the point of contention, because you cannot possibly run the test the same exact way each time as a human. In other words, he's changing the population he's taking a sample from with each test. Which makes no sense if his goal was to quantify relative performance. BUT, it really isn't, it's to give us a general idea of performance, which is the part I like.

Which is why I said I think both reviews have their advantages, one is better at gauging relative performance between cards, while the other will give you a better idea of what your actual gameplay will look like.

 

[ShadowOfMyself]

Interesting the number of new members posting here

Ive been out of the loop since the 8800GTX launch because nothing exciting has happened yet (shame on you AMD and Nvidia) but I remember reading some dreadful Hocp reviews and discussing in their forums... Wait, did I say discussing? I didnt have time to do any of that because as soon as I expressed my disappointment towards Kyle he banned me :|

He needs to shut his trap, apples to apples benchmarks are where its at... I dont care about his methods, their benchmarks are just unnacceptable... When you wanna compare 2 cars by test driving them, are you gonna use different tracks? I dont think so! Wouldnt make much sense now, would it? It might or might not be the same as "real world experience" but the lack of variables make it a good benchmark either way... Meanwhile, Kyle happily goes through a light area with the card he likes, and then when benchmarking the other card, taxes it as much as he can... Oh but.. Its real world benchmarking :roll:

 

[Rusin]

Originally posted by: madh83

You raised good points there . If this OCP way of testing is used they should deliver more information (those statistics) for perhaps like downloading so everyone could see them. This shouldn't be too big issue to make.

It would be perfect though if time demos would mirror real life performance..it would solve so many problems.
---
There are other sites like Muropaketti which did similar type of testing, but they use more of this repeating. They also mention if there was anything special happening that didn't show up on scores (like if there were sometimes very suprising lows that doesn't show up on that minimum bar). They say they get pretty similar results on each runs when they try to make it every time the same way..no significant jumps or lows.

 

[SPARTAN VI]

Originally posted by: ShadowOfMyself
When you wanna compare 2 cars by test driving them, are you gonna use different tracks? I dont think so! Wouldnt make much sense now, would it? It might or might not be the same as "real world experience" but the lack of variables make it a good benchmark either way... Meanwhile, Kyle happily goes through a light area with the card he likes, and then when benchmarking the other card, taxes it as much as he can... Oh but.. Its real world benchmarking :roll:

Allow me to modify your car analogy to fit with what Hard|OCP is doing. The video card is to a vehicle as a video game is to the track. Each car is given a target lap time (frame rate) and they both try to get as close as possible to this time. However, the only variable that can be manipulated are the settings (the settings on video game, not the entire track). So no, they're not changing tracks, rather they're changing the track's variables. In this analogy, it would be like changing the weather conditions of track (hot/cold pavement) or perhaps adding an additional lap.

This tampering most definitely could skew results in favor of which ever architecture is more fit to deal with the new obstacles. Choosing the tracks is another matter (like selecting a save game or a game in general). It would be like lining up a Shelby Mustang against an Lancer Evo on a course ridden with twisties and turns then lining them up again on a drag strip. These cards also seem to favor certain games, sometimes the 3870 X2 pulls ahead of the 8800GTX in H|OCP's evaluation (IIRC).

Feel free to elaborate people. I see potential for error and bias in H|OCP's methods, but I'm grateful that they're not performing the same reviews as everyone else.

 

[Zap]

Originally posted by: taltamir
I do like reading various sources though

Me too.

Originally posted by: smut
I think people should look at all methods, not just one and base their decision on that because each method does have pros/cons.

Exactly. For anyone to say that one method is "the only" way is being shortsighted at best.

Originally posted by: apoppin
his famed "real world" testing is just to make his site different:

That's okay. After all, if all sites were identical to each other, then the intarweb would be a boring place indeed.

Originally posted by: apoppin
i never ran across a tech savvy beggar on the street ... it appears they are all on the 'net

Of course. They're tech savvy, after all. When I first moved to the Midwest and was living out of a hotel and suitcase with no internet connection for a couple weeks, I resorted to using public libraries for internet access. It works and is free.

Originally posted by: Endgame124
Back in the Quake vs Quack days, it seemed the video card companies were cheating the benchmarks used. A video card could get a drastically better score in the timedemo than in actual gameply. Isn't that the same thing thats going on here? If timedemos were useless due to cheating 7 years ago why are people still using them today?

You can make your own timedemos, can't you? If a video card company made enough optimizations for all the different timedemos that anyone can possibly benchmark using, then effectively they've optimized for the whole game. That isn't necessarily a bad thing.

Originally posted by: keysplayr2003
Just don't run over and register at H just because of this and start anything.

I've been registered over there for years.

I never base a purchasing decision solely on one review, no matter what site.

 

[Endgame124]

Originally posted by: ShadowOfMyself
Interesting the number of new members posting here

Ive been out of the loop since the 8800GTX launch because nothing exciting has happened yet (shame on you AMD and Nvidia) but I remember reading some dreadful Hocp reviews and discussing in their forums... Wait, did I say discussing? I didnt have time to do any of that because as soon as I expressed my disappointment towards Kyle he banned me :|

He needs to shut his trap, apples to apples benchmarks are where its at... I dont care about his methods, their benchmarks are just unnacceptable... When you wanna compare 2 cars by test driving them, are you gonna use different tracks? I dont think so! Wouldnt make much sense now, would it? It might or might not be the same as "real world experience" but the lack of variables make it a good benchmark either way... Meanwhile, Kyle happily goes through a light area with the card he likes, and then when benchmarking the other card, taxes it as much as he can... Oh but.. Its real world benchmarking :roll:

But has anyone proven that timedemos play relative to actual game performance?

ex: Card A, Timedemo A: 50fps

While actually playing the game will it get 50fps?

Even worse, has anyone proven the following doesn't occur:

Card A, Timedemo A: 50fps
Card A, in game fps: 22fps

Card B, Timedemo A: 45fps
Card B, in game fps: 40fps

If you only run timedemos for Game A, card A will appear to the card to buy. But if you play the game, obviously Card B is the card to get. You may not be able to identically repeast tests while actually in the game, but knowing what you're going to get while playing the game seems like what a reader would want to know.

I'd imagine readers of any review site want to know which card actually plays the best. If the timedemo is relational to real gameplay, then use timedemos (and please provide proof that the timedemo is relational). If the timedemo is not relational, please do in game testing (and provide proof that the timedemo is NOT relational to real game performance).

Sounds pretty simple to me....

 

[ShadowOfMyself]

Originally posted by: Endgame124

Originally posted by: ShadowOfMyself
Interesting the number of new members posting here

Ive been out of the loop since the 8800GTX launch because nothing exciting has happened yet (shame on you AMD and Nvidia) but I remember reading some dreadful Hocp reviews and discussing in their forums... Wait, did I say discussing? I didnt have time to do any of that because as soon as I expressed my disappointment towards Kyle he banned me :|

He needs to shut his trap, apples to apples benchmarks are where its at... I dont care about his methods, their benchmarks are just unnacceptable... When you wanna compare 2 cars by test driving them, are you gonna use different tracks? I dont think so! Wouldnt make much sense now, would it? It might or might not be the same as "real world experience" but the lack of variables make it a good benchmark either way... Meanwhile, Kyle happily goes through a light area with the card he likes, and then when benchmarking the other card, taxes it as much as he can... Oh but.. Its real world benchmarking :roll:

But has anyone proven that timedemos play relative to actual game performance?

ex: Card A, Timedemo A: 50fps

While actually playing the game will it get 50fps?

Even worse, has anyone proven the following doesn't occur:

Card A, Timedemo A: 50fps
Card A, in game fps: 22fps

Card B, Timedemo A: 45fps
Card B, in game fps: 40fps

If you only run timedemos for Game A, card A will appear to the card to buy. But if you play the game, obviously Card B is the card to get. You may not be able to identically repeast tests while actually in the game, but knowing what you're going to get while playing the game seems like what a reader would want to know.

I'd imagine readers of any review site want to know which card actually plays the best. If the timedemo is relational to real gameplay, then use timedemos (and please provide proof that the timedemo is relational). If the timedemo is not relational, please do in game testing (and provide proof that the timedemo is NOT relational to real game performance).

Sounds pretty simple to me....

Heres is why I dont think it works that way:

When testing either a timedemo or a game segment, you are testing, well... A segment!
Do you hold more value to a random segment of the game instead of a timedemo? Why would it be more important?

The point Im trying to make is, for "real world testing" to be 100% accurate, a reviewer would need to go over the whole game with fraps on and make an average at the end, and only then you could say one card offers the better experience

By starting up a game or loading a random save HardOCP proves nothing, they are simply testing a single segment of the game... Maybe a 3870 is faster in save 4, but the 8800 is faster in timedemo 2... Maybe the 3870 is faster in timedemo 1, but the 8800 is faster in save 7... See what I mean?

Because of this, apples to apples is always better

 

[Mana]

This benchmarking the benchmarks only manages to reinforce that the timedemos that Anandtech and other websites use are relevant.

In Hardocp's real time testing the Radeon 3870 X2 is 85% as fast as the 8800 GTX. Looking at the built in Island GPU time demo comparison, the Radeon 3870 X2 is 88% as fast as the 8800 GTX. Finally, looking at their comparison with a custom timedemo the Radeon 3870 X2 is 91% as fast as the 8800 GTX.

These results, I argue, are all well within a reasonable margin of error. In other words, while the framerates seen in the timedemos are not the framerates we will see while playing the game, timedemoes still accurately show how one videocard performs relative to another videocard.

Basically Kyle just validated Anandtech's method.

 

[Duwelon]

I wonder if AT's entry in the PSU benchmarking is what chapped Hardocp's Hide. When I read AT's intro to PSU benchmarking it never singled out any specific site, but I wonder if Kyle or whoever at HardOCP felt that way.

 

[Rusin]

Originally posted by: ShadowOfMyself


Heres is why I dont think it works that way:

When testing either a timedemo or a game segment, you are testing, well... A segment!
Do you hold more value to a random segment of the game instead of a timedemo? Why would it be more important?

The point Im trying to make is, for "real world testing" to be 100% accurate, a reviewer would need to go over the whole game with fraps on and make an average at the end, and only then you could say one card offers the better experience

By starting up a game or loading a random save HardOCP proves nothing, they are simply testing a single segment of the game... Maybe a 3870 is faster in save 4, but the 8800 is faster in timedemo 2... Maybe the 3870 is faster in timedemo 1, but the 8800 is faster in save 7... See what I mean?

Because of this, apples to apples is always better

First of all time demos doesn't play game from start to finish so in real world testing you don't have to play from start to finish to see better results.

That last paragraph was confusing; you didn't make any arguments to support it, but nevertheless you made that claim. If you use same settings that problem that you implement in second last paragraph applies to both time demos and real world testing.
---

Thing is that you can try to replicate time demo by playing the same segment that time demo is playing. If you can make rendering situation easier in actual game play (and succeed in that) and still time demo would win you would know theres something wrong with time demo. I did this already.. I did try to make actual gameplay run faster than time demo, but didn't work..time demo was faster [In time demo there are heavy battle scenes, in my game play there wasn't anything much happening on screen(used same map, same settings and same amount of bots running around there)]

 

[myocardia]

Originally posted by: Anonymous Coward
Came here through Slashdot-Hard-Anand and after reading some of the comments I just can't help myself. how about you Apoppin tell if the results they found is somehow faulty and don't represent the real game on stop masturbating on those saves.

So, should we assume this was Kyle, because of the Engrish, along with the hate for apoppin? It seems so, doesn't it?

Originally posted by: Rusin
One of my points is that Kyle ain't only one getting these results with same type of testing method.

You know, this is about the third time you've said that, yet you've provided no links. Care to make us not think you're either outright lying, or talking about what was posted on some fanboy site? If so, here's your chance.

Originally posted by: McCartney
I feel like the guy everyone called crazy, but I'm the genius.

Don't worry, that's definitely not the case here.

Originally posted by: ShadowOfMyself
Interesting the number of new members posting here

It is, isn't it?

He needs to shut his trap, apples to apples benchmarks are where its at... I dont care about his methods, their benchmarks are just unnacceptable... When you wanna compare 2 cars by test driving them, are you gonna use different tracks? I dont think so! Wouldnt make much sense now, would it? It might or might not be the same as "real world experience" but the lack of variables make it a good benchmark either way... Meanwhile, Kyle happily goes through a light area with the card he likes, and then when benchmarking the other card, taxes it as much as he can... Oh but.. Its real world benchmarking

Oh, finally, the truth rears it's ugly head. Thanks for the heads up, Shadow.

 

[schneiderguy]

Originally posted by: Mana

In Hardocp's real time testing the Radeon 3870 X2 is 85% as fast as the 8800 GTX. Looking at the built in Island GPU time demo comparison, the Radeon 3870 X2 is 88% as fast as the 8800 GTX. Finally, looking at their comparison with a custom timedemo the Radeon 3870 X2 is 91% as fast as the 8800 GTX.

These results, I argue, are all well within a reasonable margin of error. In other words, while the framerates seen in the timedemos are not the framerates we will see while playing the game, timedemoes still accurately show how one videocard performs relative to another videocard.

:thumbsup: 5% is well within the margin of error especially when doing "real world" tests for one of them that can't be reproduced exactly.

HardOCP just likes to hide behind their unscientific method of "benchmarking" so they can paint the picture however they want (or how their sponsorers want).

 

[SickBeast]

I personally *hate* the videocard review format at HardOCP. I stopped visiting their site on a regular basis ever since they stopped using "apples-to-apples" comparisons.

I find their layout terrible. Their tiny thumbnail images are unintuitive, and they do not have the 'mouse-over' feature that the AT articles use. It seems like they're making un-necessary changes to their review-strucuture that actually detract from what the reader learns from their articles.

The way they have split the site makes no sense. I personally am interested in all sorts of hardware news; I find myself scratching my head whenever I go to HardOCP for news. Am I missing out on something? Do I need to read the "Enthusiast" section to find out about what I'm looking for?

Back in the day, there was pretty much AnandTech, HardOCP, and Tom's Hardware. AT has clearly moved to the forefront of the three sites. Tom's is no longer owned by Tom, and has basically become another Cnet. HardOCP has tanked for the reasons I outlined above.

As others have said, their insults toward AT are based on jealousy, a feeling of inferiority, and a desperate means of attracting traffic to their site.

Kyle knows a great deal about hardware, but I've got to say that I'm disappointed in his site over the past few years. I remember reading his site about 3 months before the Phenom was to be released, and he was saying how excited he was about it after a hands-on test. Phenom was such a disappointment! This all has really made me think twice before getting information from them!

 

[Rusin]

myocardia:
http://plaza.fi/muropaketti/ar...deon-hd-3870-x2-r680,2

This site also attacked against Anandtech's testing methods. This site uses also actual gameplay, but uses same settings for each card.

It's funny how people are talking about that running time demos would be "scientific method".

As said in this one uknown forum:
"The point of the scientific method is to test a theory and prove it in the real world".

 

[CP5670]

Kyle knows a great deal about hardware, but I've got to say that I'm disappointed in his site over the past few years. I remember reading his site about 3 months before the Phenom was to be released, and he was saying how excited he was about it after a hands-on test. Phenom was such a disappointment! This all has really made me think twice before getting information from them!

If I remember correctly, these were also the same guys saying that the C2Ds didn't give much improvement in practice and were not worth buying, back when those first came out. :roll:

 

[apoppin]

Originally posted by: Rusin

Originally posted by: apoppin

When Kyle benches you don't know ANY variables he uses
When AT benches you know it is a "clean" bare-bones rig set up from a image and all the games are played at optimum conditions for gaming

One of my points is that Kyle ain't only one getting these results with same type of testing method.

i have no problem with that. The one thing you will notice that all these "real world" testing sites have in common - their results are not consistent.

-I deliberately did try to avoid the most graphically hectic moments.. clear difference to that time demo.

and that will account for most of the difference

What? Please explain? I mean with that actual game play I didn't have much going on that screen unlike in that time demo. Still that actual game play lost. Are you really claiming that if there were more happening (more things to render) on screen it would have ran faster?

Oh.. I also tested that (as mentioned in that post) and I was consistently getting 20 fps smaller fps. Yes..difference actually got bigger..not smaller as you did predict.
---

I deliberately did try to make actual game play to have better numbers, but I failed. I did for sake of my argument: I'm saying that these time demos run faster..so I did give actual game play better grounds to work.. but still time demo were faster.
---

Here is that test:
http://downloads.guru3d.com/download.php?det=1816

I heard that recent driver update would give Radeon users chance to use AA in Unreal3 based games..you could try that.

Thanks for the link. the only problem is that i do not have UT3
... so the benchmark tool would do me no good whatsoever. Are you saying you covered the *exact same path* in your map as in the demo?
*any variation* will change the results ... how close were your runs in terms of Min/Max/Average?

Finally, i never even saw the UT3 benchmark [remember i said i only bench games i am interested in and actually PLAY] so this is a 'stab in the dark' - perhaps the AI and physics are different in your run and in the timedemo.

========================

Originally posted by: CP5670

we already do that ... it is SO easy to make a custom time demo; Keys made a good one for STALKER - the USUAL thing is to POST it for everyone to replicate and for peer review. Kyle's is HIDDEN ... no one can review his benchmarks because we don't know which section of the game he choose.
:roll:

unethical is a word that springs into my mind

... and there is NO way to keep recorded timedeomos from "falling into the 'hands" of the GPU companies ... but it is SILLY for them to optimize for each benchmark - they really need to optimize the game ... and generally they do just that

I don't see how they could get the demos if they aren't published, unless they have a shill among the review staff. It's not ideal to keep them hidden, and the site would need to have some credibility already established (which HardOCP frankly does not), but this would prevent the companies from optimizing specifically for the demos. This situation is still better than having actual gameplay runs that even HardOCP admits can never be the same between different cards.

No. IF i want to bench FarCry for example, i have a *library* of established benchmarks to choose from. i can D/L them, you can d/l them and Derek already has the collection plus probably some custom ones of his own. These "established" and "peer reviewed" benchmarks are available for most anyone - including nvidia and AMD. BUT, i am saying it is SILLY to just "optimize for specific benchmarks" - IF they were to get caught they would have their credibility shot to hell. And HardOCP has not come close to "catching them" in any 'conspiracy' to defraud us. Otoh, we are close to unmasking HardOCP and Kyle's motives.

Now if i make a "custom timedemo" and refuse to share it for peer review and "claim" that GPU A is 50% faster than GPU B on THAT basis, then everyone has every RIGHT to DOUBT my benchmark. What is what Kyle and HardOCP are doing.

again to KyleB ....
SHOW us the Crysis video of your run ... give us your Crysis 'save' so we can *confirm* your results and lay this crap to rest - the crap YOU brought up. PROVE it. i triple dare you to even try.

---------------------------------

Originally posted by: McCartney

Originally posted by: apoppin
thanks ... and welcome

i sent a PM to Derek ...


i sense a war brewing ...

i do have to agree with their intro, however:

HardOCP has been thoroughly mired ...

that is the correct term

HARDFUD eh?!

LEGIT BENCHMARKS>

I feel like the guy everyone called crazy, but I'm the genius. Seriuosly apoppin man, why did you have to try to insult my favorite forum? I love AT, but you just had to stab at hardocp. Your beloved review is falsified, and the 3870x2 was never king of the hill. Sorry man.

you ARE a genius ... you are the one who ultimately will be known as the 'cause' of the complete public humiliation and breakdown of your site's credibility ...


good one

make sure you pass my message to KyleB ... give him my best regards ... and please remind him that we'd love to see and test his Crysis Demo run for ourselves