OT: Website suggestion

[polobunny]

Originally Posted by Capper

Every conversation turns into this.....and frankly it gets old.....90% of your posts are argumentative. With the fact that you know everything, and know better than anyone here....put your nuts out there and start doing some reviews....I'm sure with your knowledge all these companies would throw money at you, and you could retire from your current job and just put out 2-3 articles a week.

If only.

No need to get on my back though.

[Capper]

Look, Yves....you are a good guy, I like you.....but I'm not the only one who has noticed that you really like to argue everything.....

I'm tired, I'm grumpy, and you hit a nerve with me. I could sit here all night explaining how and why, and you'd still have to argue, and it pisses me off to no end sometimes.

[MantaBase]

Hmmm... I think Cappy needs a rest.. He has been working too hard!

Look, it comes down to this....

There are better ways to review a video card. The problem is that most of them are very impracticle and add complication that most folks can't assimilate. You (Polo) would like to see systems reviewed that are similar to the ones you use. If it takes 8 hours to review a card, you just doubled it to 16. If it takes 24, you just made it ~50. This is one of the reasons we link to other reviews when we do ours. NO REVIEW is all knowing and comprehensive. I personally would like to spend ~ 30 hours on the testing of each PSU I review. But, that's not practical. As it is, we are buried in PSUs and our test suite takes ~4 hours from start to finish. That doesn't include the 4-8 hours required to research and write the review.

Further, what Rich pointed out an you may not get is that when you do a video card bench, you want to test the VIDEO CARD, not the system. That means you want to remove as many bottle necks as possible. I understand your issue - it's not lost on me. However, if you use a system in which the CPU is actually bottle necking your results, then you are not benching the video card anymore - but the system (specifically the CPU). In a video card review, that's not very helpful.

I fully understand your desire to see what SYSTEMS bench at. That would be very useful. It's also very impracticable to test. Which is why, as good as an idea as it may sound, you don't see many doing it (in fact, I don't see any doing it).

Does that make sense? If you have some uber idea that all the review site have somehow missed to get around this - I'm sure we would love to hear it.

What reviewers would love is a computer with dials we could turn up and turn down - Alas, no-one makes one.

[polobunny]

Originally Posted by Capper

Look, Yves....you are a good guy, I like you.....but I'm not the only one who has noticed that you really like to argue everything.....

I'm tired, I'm grumpy, and you hit a nerve with me. I could sit here all night explaining how and why, and you'd still have to argue, and it pisses me off to no end

I'm not too sure where you see me arguing in this thread Cap.

First post I'm telling about my gripe about how most scores are with computers that are way too expensive for me. How I would like to see some scores with a bit more normal computers, maybe ones I could afford (hey that would be neat) so I know what to expect if I buy X component.

You tell me about how you don't want any component bottlenecking when testing a certain component, which I perfectly understand. If you want to get the value of a component alone, you obviously don't want the rest of the system to be a drag and slow it down.

Now for the sake of it, I understand you and anyone else who reviews components can't test one component with a multitude of systems, but there's this trend going in most if not all reviews I read where the test bed is far from mainstream.
How am I supposed, as a guy with only a largely mainstream computer, wanting to buy a certain component to know how performance in X application would scale if I bought that component?
The only way I would be sure is if I had one of these crazy ass computers...


Edit: Manta, then maybe someone has to take the necessary time to test on more mainstream parts to make it easier for the majority of users to see what they should expect.

[Capper]

I do need a rest, but unfortunately I have too much stuff that needs to be done.

I have no problem with questions, or a comment here or there.......but when it turns into questioning/criticizing almost constantly, it wears thin. This was the issue i had last week with the person in the thermal compound thread.......all logic went out the window and it got to the point where it was 'Did you measure each thermal compound to make sure you were using precisely the exact same amount each time?"......seriously.

I think our reviews are extremely accurate, and well done....I take a lot of pride in the time and effort that goes into everything we do, and I think our results are validated by the results others see. As far as the test bench, its impossible to have several different parts on hand, especially when we give almost everything away, and stuff like processors come out of pocket.

Having said that, almost all of my reviews have been done with an E6850 and an 80GB SATA II HDD. Hardly what I'd call cutting edge. I did recently switch to an 8800GTX, but that card died and its back to an 8800GTS.

I do use a QX6850 in some reviews, for the fact that it runs hotter than most any processor, and the unlocked multiplier allows me to emulate most any processor.

[Stormcrow]

When it comes down to it, it's the card itself that only needs to be benched and not the entire system.

Lets say you have an 8800GTS, 8800GT, HD3870, HD 3850, and so on down to like a 8500GT. You want to test those in a system that removes any possible hindrances because as Manta said, otherwise you'll be benching the system itself and not just the card's performance. The point is even if the review has a small god-like uber rig and you have a more mid-range setup, the final verdict of the card doesn't change. The 8800GT is a better card then the HD3850, if you have the budget get that. If not, the HD3850 is better then the 8800GS, which is better then the 8600GTS, and so on. Getting a real world analysis for the average enthusiast setup is not going to change the fact the 8800GT is better then a HD3850, the point is we know this for a fact as the test setup didn't hold back the 8800 which would skew the results making the HD3850 seem more competitive. If you want the better card, regardless of system, the 8800GT is your best pick. You don't look at how many actual FPS the card is pulling, but the ratio of FPS that card has compared to other cards in the same system, the frames per second themselves are irrelevant to deciding which card outperforms which but are the best measure for in game performance.

I hope I kinda make sense.

[MantaBase]

Originally Posted by polobunny

Edit: Manta, then maybe someone has to take the necessary time to test on more mainstream parts to make it easier for the majority of users to see what they should expect.

OK, Lay it out for me - and remember, a review must be for the masses - not just Polobunny :)

We are always open to ideas - we just have this "reality" thing that keeps pounding on our heads.

[Capper]

And I apologize for being so short with everything....I am very tired and haven't been feeling well

[Stormcrow]

Go to bed Rich, take some time to relax a little.
I'll have some articles for ya in the morning

[Capper]

Well, my night got really exciting a bit ago......the police helicopter was flying overhead, I look =out the back window and some black guy is scaling our wall. Went downstairs to intervene, and he had already made it around front and was headed up the street.....now the Wife and dog are both scared.

I'm trying to get about 5 things done at once, line up articles, and help people......IO couldn't imagine not having the guys i have helping out.