Value of Hardware Unboxed benchmarking

[Silent_Buddha]

They review products and features, not sides.

Maybe the perspective that the reviews are about vendors or "sides" could just be a projection that some people make because that's how they view these things for whatever reason?

Imagine that. Not everyone on the planet thinks in exactly the same way. It's almost like they are all different people with different views and values and perceptions. And almost like opinions can change based on a changing marketplace (like more games actually having DLSS now compared to before and DLSS being better now compared to before, albeit still not good enough for me).

Regards,
SB

 

[Kaotik]

Imagine that. Not everyone on the planet thinks in exactly the same way. It's almost like they are all different people with different views and values and perceptions. And almost like opinions can change based on a changing marketplace (like more games actually having DLSS now compared to before and DLSS being better now compared to before, albeit still not good enough for me).

Regards,
SB

Yet some not being able to accept that led to closing one of the subforums.

 

[trinibwoy]

Imagine that. Not everyone on the planet thinks in exactly the same way. It's almost like they are all different people with different views and values and perceptions. And almost like opinions can change based on a changing marketplace (like more games actually having DLSS now compared to before and DLSS being better now compared to before, albeit still not good enough for me).

Regards,
SB

That’s true but the review scene has changed a lot over time. Back in the day the most trusted reviewers were very cautious about sharing subjective opinions or making predictions of the future. They stuck to facts and technology. M The most popular reviewers today are a lot more bombastic and reviews have become opinion pieces for entertainment value. So while everyone is entitled to an opinion some opinions have more influence on the market than others. E.g. random forum dude vs YouTuber with a million subscribers.

 

[troyan]

There is a whole hate youtube industry which can live from making videos about nVidia. Channels like this one asking for patreons and cant be objective anymore.

 

[Silent_Buddha]

That’s true but the review scene has changed a lot over time. Back in the day the most trusted reviewers were very cautious about sharing subjective opinions or making predictions of the future. They stuck to facts and technology. M The most popular reviewers today are a lot more bombastic and reviews have become opinion pieces for entertainment value. So while everyone is entitled to an opinion some opinions have more influence on the market than others. E.g. random forum dude vs YouTuber with a million subscribers.

Sure, but even back when the FX 5800 launched, some sites lauded it as better than the 9700 Pro due to it performing well in DX 7 and 8 titles. Nevermind that it was rubbish at DX 9. But then you could make a case for that if you primarily played DX 7 and 8 titles.

So in the vast majority of titles at the time, DLSS was a non-starter for many as it was limited to just a few games and its implementation wasn't great. To me, it's obvious at the time that for anything that mattered to me, the AMD cards (RDNA2 cards that is, RDNA was somewhat competitive but not compelling for me) were competitive and superior in many cases even if DLSS and RT would eventually, perhaps, become a compelling feature in more and more titles.

You can also see the same thing back in the day with NV cards being crowned by some sites despite having subpar AA quality compared to ATi cards. Because for some people, the quality of AA didn't matter and some even ran without AA due to the performance impact.

Hell, there's been times when sites didn't care that NV had worse AF that was incredibly angle dependent and instead crowned as better at AF purely due to performance even though the quality was truly atrocious, IMO. They eventually made it better than ATi/AMD's implemention but it took a while.

So, I really fail to see any material difference between Hardware Unboxed reviewing things as they have versus many of those sites in the past. While I didn't agree with them, they certainly made valid points for some people that didn't value the same things as I did.

Regards,
SB

 

[Kaotik]

Hell, there's been times when sites didn't care that NV had worse AF that was incredibly angle dependent and instead crowned as better at AF purely due to performance even though the quality was truly atrocious, IMO. They eventually made it better than ATi/AMD's implemention but it took a while.

Regards,
SB

This went back and forth, but NVIDIA was the last angle dependent when AMD and Intel had angle independent solutions IIRC

 

[trinibwoy]

Sure, but even back when the FX 5800 launched, some sites lauded it as better than the 9700 Pro due to it performing well in DX 7 and 8 titles. Nevermind that it was rubbish at DX 9. But then you could make a case for that if you primarily played DX 7 and 8 titles.

So in the vast majority of titles at the time, DLSS was a non-starter for many as it was limited to just a few games and its implementation wasn't great. To me, it's obvious at the time that for anything that mattered to me, the AMD cards (RDNA2 cards that is, RDNA was somewhat competitive but not compelling for me) were competitive and superior in many cases even if DLSS and RT would eventually, perhaps, become a compelling feature in more and more titles.

You can also see the same thing back in the day with NV cards being crowned by some sites despite having subpar AA quality compared to ATi cards. Because for some people, the quality of AA didn't matter and some even ran without AA due to the performance impact.

Hell, there's been times when sites didn't care that NV had worse AF that was incredibly angle dependent and instead crowned as better at AF purely due to performance even though the quality was truly atrocious, IMO. They eventually made it better than ATi/AMD's implemention but it took a while.

So, I really fail to see any material difference between Hardware Unboxed reviewing things as they have versus many of those sites in the past. While I didn't agree with them, they certainly made valid points for some people that didn't value the same things as I did.

Regards,
SB

I have a different recollection of that history. Anandtech and Tomshardware were staples back then and I recall their coverage of all of those issues (FX performance and noise, AA, AF, benchmark cheating) to be very balanced and analytical without lots of bluster. I have fond memories of the AF flowers and AA pixel grids that littered their reviews. I certainly don't recall anyone respectable hailing FX 5800 as a world beater vs the 9700 pro.

HWUB claims that they make recommendations based on things they can test today which is perfectly fine. So from that perspective their skepticism of any forward looking tech is somewhat justified. But some of their takes are just downright silly. In their Turing launch review for example they highlighted that "only" 1/4 of the big games launching in the following year may get RT and people probably won't play most of them anyway.

It was a pretty stupid comment because any logical person understands you're not going to get a ton (or any) games supporting new hardware features on day one. That's not how the industry works and it usually takes many years for features to be adopted (mesh shaders anyone?). Of course HWUB knows this yet they chose to play up this negative spin anyway. To add to the silliness they also claimed Shadow of the Tomb Raider getting an RT patch a month after launch was no big deal because most people interested in the game would've already played it by then anyway. That's clearly not true. Skepticism is fine but they were very negative out the gate for no good reason.

 

[techuse]

I have a different recollection of that history. Anandtech and Tomshardware were staples back then and I recall their coverage of all of those issues (FX performance and noise, AA, AF, benchmark cheating) to be very balanced and analytical without lots of bluster. I have fond memories of the AF flowers and AA pixel grids that littered their reviews. I certainly don't recall anyone respectable hailing FX 5800 as a world beater vs the 9700 pro.

HWUB claims that they make recommendations based on things they can test today which is perfectly fine. So from that perspective their skepticism of any forward looking tech is somewhat justified. But some of their takes are just downright silly. In their Turing launch review for example they highlighted that "only" 1/4 of the big games launching in the following year may get RT and people probably won't play most of them anyway.

It was a pretty stupid comment because any logical person understands you're not going to get a ton (or any) games supporting new hardware features on day one. That's not how the industry works and it usually takes many years for features to be adopted (mesh shaders anyone?). Of course HWUB knows this yet they chose to play up this negative spin anyway. To add to the silliness they also claimed Shadow of the Tomb Raider getting an RT patch a month after launch was no big deal because most people interested in the game would've already played it by then anyway. That's clearly not true. Skepticism is fine but they were very negative out the gate for no good reason.

These statements were made in the context of trying to find value over the existing Pascal line considering the lack of any performance/price increases. The only differentiator was these features. The only
GPU that offered a performance increase was the $1200 2080ti. A 70% price increase over the 1080ti for 30% more performance. Those early years of RT were a complete bust honestly. Performance was too low on those GPUs, implementations sucked and DLSS was completely useless.

 

[Silent_Buddha]

I have a different recollection of that history. Anandtech and Tomshardware were staples back then and I recall their coverage of all of those issues (FX performance and noise, AA, AF, benchmark cheating) to be very balanced and analytical without lots of bluster. I have fond memories of the AF flowers and AA pixel grids that littered their reviews. I certainly don't recall anyone respectable hailing FX 5800 as a world beater vs the 9700 pro.

Yeah, those were some good ones. As was TechReport (one of the very few sites to point out that even though the HD 2900 XT wasn't a great card, it still brought some nice advancements to AA). I even enjoyed Hard|OCP's attempt at "real world" comparisons to an extent. But there were a lot of review sites that I used to keep and eye on, many of which aren't around anymore, some of which still are.

Regards,
SB

 

[trinibwoy]

Techreport was a big loss. So was Anandtech. Times have definitely changed.

 

[Florin]

They review products and features, not sides.

Maybe the perspective that the reviews are about vendors or "sides" could just be a projection that some people make because that's how they view these things for whatever reason?

That's possible, sure. Is it not at least equally possible that such perspectives have some actual justification based on examples found in the long and voluminous history of HUB's public expression?

I'm not sure that reviewers should be assumed to have no personal views when everyone else does.

 

[Putas]

Sure, but even back when the FX 5800 launched, some sites lauded it as better than the 9700 Pro due to it performing well in DX 7 and 8 titles. Nevermind that it was rubbish at DX 9. But then you could make a case for that if you primarily played DX 7 and 8 titles.

Since there were no dx9 games, it would be too much guessing for some.

 

[troyan]

Sure, but even back when the FX 5800 launched, some sites lauded it as better than the 9700 Pro due to it performing well in DX 7 and 8 titles. Nevermind that it was rubbish at DX 9. But then you could make a case for that if you primarily played DX 7 and 8 titles.

More like: Reviewers would have had recommended a CineFX card despite having the knowledge of how bad these have been with Pixelshader 2.0. This channel did it with RDNA2 and Ampere.

/edit: This channel would have benched games with Pixelshader 2.0 support only with DX8 and Pixelshader 1.4, because the improvement in image quality wouldnt have been worth it and most gamers wouldnt have cared about DX9 anyway.

 

[davis.anthony]

Maybe the perspective that the reviews are about vendors or "sides" could just be a projection that some people make because that's how they view these things for whatever reason?

Or maybe, it's just their reviews tend to tip slightly to one side more than the other.

 

[Silent_Buddha]

Since there were no dx9 games, it would be too much guessing for some.

Eh, FX 5800 launched in March, 2003. There were already a few Dx9 games out when the FX 5800 launched. The biggest was probably Freelancer which came out a few days before the card was released.

Regards,
SB

 

[Putas]

Eh, FX 5800 launched in March, 2003. There were already a few Dx9 games out when the FX 5800 launched. The biggest was probably Freelancer which came out a few days before the card was released.

Regards,
SB

Thanks for the correction. Did anyone benchmark it?

 

[Phantom88]

Ready For The Winter Games: ATI Radeon 9800 XT

Freelancer was released by Microsoft early this year. It's 3D engine is a bit outdated but the game has a big fan community. It is a proxy of games using less advanced graphics technology. The framera

www.tomshardware.com

Its the 5900 here, but looking at the drop for 5600, the 5800 would have probably behaved similarly.

 

[troyan]

Thanks for the correction. Did anyone benchmark it?

There were a few games with Pixelshader 2.0. Biggest hit for nVidia came with the HL2 presentation in mid 2003: https://www.anandtech.com/show/1144

 

[trinibwoy]

Ready For The Winter Games: ATI Radeon 9800 XT

Freelancer was released by Microsoft early this year. It's 3D engine is a bit outdated but the game has a big fan community. It is a proxy of games using less advanced graphics technology. The framera

www.tomshardware.com

Its the 5900 here, but looking at the drop for 5600, the 5800 would have probably behaved similarly.

Wow which site was doing 1% lows back then or are those recent benchmarks?

Any reviewer worth their salt came to the same conclusion on the 9700 Pro vs FX 5800 even without benchmarking a lot of DX9 games because they understood the tech and also understood programmable shading was the future. Tom's summarized the DirectX9 situation nicely.

Only very few DirectX 9 games have hit the market these days, so judging the performance of a card under DX9 is not an easy task right now. We can currently see one trend: NVIDIA's FX cards are in trouble. The reasons are manifold. The code that comes out of Microsoft's HLSL (High Level Shader Language) compiler does not go very well with NVIDIA's CineFX architecture and the cards do not offer 24 bit floating-point performance. They have to use the much more resource hungry 32 bit mode instead or fall back to 16 bit, which results in a loss of image quality in most situations.

ATI's Radeon 9500/9600/9700/9800 can run the code that comes out of the MS HLSL compiler - the so called DX9 standard code - very well and they also use the best compromise in terms of floating point quality by using the minimum DX9 spec 24bit. It seems that the whole design of the ATI Radeon DX9 chips was designed to fit to HLSL code. If you intend to buy a new graphics card you see that ATI's DX9 cards run fast and without trouble in actual DirectX 9 games, while NVIDIA's FX cards rely on optimized code paths or driver optimizations.

 

[Phantom88]

They're modern benchmarks. Modern look at the cards of the era, with mature drivers and better cpus

Also, found this site some time ago, some nice benchmarks of older games not very common

Hardware museum

hw-museum.cz

Even has for Undying, back in 2001, one of the best fps games i've ever played and which has better rain and weather effects than Linneman's darling from that year, mgs 2