The Way its Meant to be Reviewed?

Hey Volt, no I think you're right. Although, I don't know that I personally see 3DMark03 as a very accurate game benchmark. AM3 is another story though and fits into that slot a bit more easily IMO. The whole PS inclusion (or lack thereof rather) in 3DM03 and AM3 caused me some confusion as to why so little was actually in those "future" benchmark offerings. Any links to articles that may explain this are more than welcome.

 

Hubba-wa? What game is 3dm2k3 benching? It's the synthetic to end all synthetics, to the best of my knowledge.

AM3 is a game benchmark, but I really think more people bench it than play the game....

 

Let me reiterate my previous post.

Correct me if I'm wrong but I thought 3DMark03 is not considered a synthetic benchmark (by NVIDIA) anymore, but rather a game benchmark. Same applies to AM3, but Massive seemed to adopt the term (game benchmark) at the launch.

 

Wasn't the "synthetic game benchmark" thing coined by NVidia to explain why they are allowed to include driver optimisations for 3DMark even though it goes against their self-avowed guidelines?

Personally, I think the DX9 demos provided by the big IHVs (i.e. Dawn, Chimp etc) are the kind of things that would be very useful. They are designed to push the new hardware and look very good. The problem here is that it's not in the best interests of these IHVs to include any real benchmarking capability because they might be accused of 'fixing' the demo to work better on their cards if they were considerably faster than competitors. On the other hand, there would always be the worry that the competitors card might be much faster and this would be very embarrassing and potentially damaging for them!

Such graphically-complex demos obviously take a lot of work both artistically and in programming, probably more than just one or two amateurs could manage to do on their own time. Didn't the old demoscene have collaborative groups of people working on different demos? We need the same sort of thing for modern graphics demos too.

 

And the answer IS:

Yup, and it's just as much BS as the 5200 being a dx9 part. 8)

 

Great point, I would like to use these more as well since they are so advanced (artistically, geometrically) in many ways over other synthetic benchmarks. You're absolutely right though, it's a classic catch 22.

 

I agree giving "Joe" too much info that he does not understand is bad. However what's worse is holding out on info that will (with a high degree of probability) make an impact on his gaming experience. Its is then the review's job to give "Joe" the info in a way that he can understand it. And it maybe as short as a few lines, saying "When we looked at this synthetic benchmark ABC we saw that card Y was faster then card Z in all cases which leads us to believe that performance in upcoming titles that use DXX on card Y will be a bit faster." Well maybe not that simple but I think can get the idea here. When other well know sights say no synthetics they are doing more harm to their readers than good. Again I am not saying Synt are the be all end all. I am saying the are an important part of the over all equation.

Not to single you out but I have seen this point argued before. If "I" don't like the game or the game is "bad" then its results don't count (again not picking on you). Well it certain counts to all of the people that own the game and have to "suffer" with it on their hardware. And last I Check one of the games was the top seller in the month of Oct. I agree their are many games that will vastly outshine these titles. But to deny the results they show is a bit of like closing your eyes to something that you don't want to see. Again not picking on you..

 

Well done article...at the end of the day though we know $$ does the talking.

 

I'm but a pup compared to you and Rev. My first review was on March 10 of this year and I recently completed my most recent (and third overall) review back in September. I don't mean to overstep my bounds here as I have a good deal of respect for your history and what you've contributed to the community. I'm always about learning more and continuing to improve my writing style, review techniques, etc.

This article particularly interested me enough to jump in the forum here to help stir the pot a bit which is why I'm here a bit "out of the blue". I believe that you can obtain valuable information on a topic if you respectfully challenge one's viewpoint. That is basically what I'm doing here. I have no agenda other than to find out more and to prove myself wrong and adjust if necessary. I take the review process very seriously as I know you all here do as well. People's hard earned money is at stake. So whether they're choice is to purchase a video card or a baby stroller; the reviews that they read on said products are obligated to be as accurate, honest and helpful as possible IMO.[/quote]

 

No offense taking at all. You're right and I basically conflict with my main goal of keeping the consumer's needs a top priority in saying that.

*begin backpeddling* ... *ok, stop...wasn't that bad*

Excellent points. See, I'm learning already.

 

Good article.

Seems rather tricky to balance a review that helps display and predict what a graphic card can or can not do...

Somewhere in the article I believe I read someone say the first generation card is for developers whereas the next cycle is for the consumer. Odd as that sounds, I think it is spot on... Also somewhat supported by Tim's view on general consumer buying cycles @ 18 - 24 months.

I mean, how long as DX9 been out now? How long as there been DX9 cards available? And...how many games support either a full suite of DX9 vs partial? And does it really matter in those games? FPS vs MMORPG, etc.

Some of the larger games expected this year, HL2 and Doom3, how much of the cards feature sets do they use? And if they are delayed, as several of the larger games are, what then?

Seems a lot of options out there - and not everyone wants the same information. You have the power users, want to be power users, game fanatics, speed junkies, and then those who want to keep up with the Jone's...

As long as the reviews can accuratly and consistantly show how well the card supports DX7,8,9, and 9.x feature sets - it is a plus. Then run tests that show unique strengths and weaknesses - maybe use the gpu manufactures demos here - you get a good overall impression of the card.

At this time the reader can choose what is important to them and make their purchase...

 

As for video cards and cars I wonder how many purchasers of high end products do not investigate the performance characteristics.

 

Right, well I've always thought the "car analogy" is about as apples-to-oranges as you can get anyway. People seem to like it though regardless.

 

In Search of an Unbiased Review

With all due respect Max Power, I do take your sites reviews with a huge grain of salt. Why? Because your an NV fan site. At heart your a fan of NV. So in your reviews your going to try and paint a better picture of the GFFX card in your article.For example,by using games that perform well with NV and avoiding the games that don't, using setting(resolution,AA/AF,etc.)that show the NV card in its best light. The same goes for Rage3D with there ATI reviews(maybe even more so ). But this is expected of a fan site. What bothers me is the sites(almost all of them)that claim to be unbiased,aren't.

I'm just an average Joe who likes to use and abuse hardware, and all i want is an HONEST opinion from the reviewer.I feel I get that here at B3D.Thanks to Wavvy and Rev.

P.S. Merry Xmas to those that observe it!

 

Re: In Search of an Unbiased Review

Not Nick of course.

 

Just jumping in and say great writing! Very good Rev.. 8)

 

Re: In Search of an Unbiased Review

Nope, the jury is still out on Nick. j/k

 

Re: In Search of an Unbiased Review

With all due respect to you as well Mr. Sparkle I believe that you're making an overly broad statement in judging a book by its cover. Simply because we are an NV fan site...is that your only reasoning? No, I am not a fan at heart of NV thank you very much....being a "fan" of a video card company is borderline lame IMO I am a fan of hardware that does what it says it will do for a good value. Period. I don't care if it is NVIDIA, ATI, etc. Help me out here Tim.

Have you actually read my reviews? If not then I'd like for you to whenever you have some time and let me know if you still feel this way. Please read them and don't just skim.

http://www.nvnews.net/reviews/gainward_gf4_ti4800SE/index.shtml
http://www.nvnews.net/reviews/bfg_geforce_fx_5900_ultra/index.shtml
http://www.nvnews.net/reviews/bfg_geforce_fx_5700_ultra/index.shtml

Also, check out our forums sometime...many of us own ATI cards as well as NVIDIA cards. We're as critical (or moreso) of NVIDIA at times.

 

Well the reason why some are so "Anti-Sythetic" is pretty obvious imo. There are an entire "specific" camp of people who are all suddenly "anti-Synthetic". We pretty much all know who and why.

In the defense of that camp there is one thing to Consider. Sythetic benchmarks may show one hardware to be weaker today. But do not take into account the very nature of the Industry today. That of Game companys partnering with IHV's. Thus say there are 20 Huge titles in a Given year. Odds are that at least half (likely more than half) will have some kind of Partnership or arrangement or Help from one or More IHVs in its development. The result is that the "weaknesses" of the hardware shown in synthetics may never apprear becuase of all the specific code developers and IHV's work together to produce to work around any porblems before the major games are released.

That is why the Synthetic issue is a tricky one. On the other hand the gamer who like to play the fringe games, or the independant games etc NEED to see the synthetics. Because they accurately show what you can expect to get from said cards without 6 months of babysitting the Game Code.