Anyone feeling disenchanted with (most) review sites?

[BFG10K]

I like BFG's, of AlienBabel, high quality tests in older games. But the reviews do suffer from a couple of Lonyo's complaints. One being he only tests a few cards at a time. It would be nice to see what a mid range card from each vendor can do in comparison to high end cards.

Heh, thanks. :thumbsup:

The limited card selection is for two reasons:

1. I simply don’t have many cards lying around to test.
2. Some people complain if you mix cards that aren’t price and/or performance competitive in the same charts.

I also purposely cut the number of games I test because again, some people complain about a large title selection, especially older games. Ironically it’s older titles running at cranked settings that often show the prowess of new hardware better than new games do.

A recent example of this is Doom 3, where the GTX480 was 32.59% faster than the GTX470 at 2560x1600 with 8xMSAA, the highest percentage gain I witnessed in my entire GTX480 performance review. You cannot tell me that old games are worthless for testing new hardware.

My priority is to test as many games as practically possible at the actual settings I play them at. This produces practical benchmark results that are founded in reality. It also produces non-uniform testing styles and patterns which are more likely to break drivers. Then I follow up later with IQ/bottlenecking/etc articles if needed.

This unorthodox style is not for everyone but it also shows a lot of stuff more traditional reviews don’t show, and that will always be my goal.

 

[Skurge]

F1 tested at 2560x1600 at 35-39 fps.... How meaningful are those results for PC racing fans? 35 fps "playable" in a racing game on a PC is an insult to anyone who plays racing games.

Once again I don't agree with this. I have a crap load of racing games. I play GTR2 and GTR evo a lot. Those 2 games run at about 170FPS, never dip below 150fps. Then I go and play Forza3 that is locked at 60fps, and its still perfectly playable. Then I go play Dirt2 or F1 at 35FPS and its still playable. Its playable cause its constant and its smooth. If it doesn't dip below 30FPs, its fine. If it jumps between 30 and 60, that's when you have problems. Its not a game by game basis thing either. I have played GTR2 on my old rigs with numbers in the 40s.

I will say I would prefer a solid 60, but you make it sound as if the game is absolutely unplayable at those rates. A game with I don't think you have yet.

 

[RussianSensation]

I will say I would prefer a solid 60, but you make it sound as if the game is absolutely unplayable at those rates. A game with I don't think you have yet.

No, I haven't grabbed the F1 game yet but I will once price drops a bit more. For example, some websites got 40-50 fps on Radeon 4890 with 4AA at 1920x1080 in Dirt 2 w/DX9. However, I couldn't play that game with my card on those settings even after overclocking it. So to me it was unplayable and choppy with 4AA despite averaging 45 fps in my own rig in the built-in benchmark. I had to reduce to 2AA or 0AA even. With 470 I don't have a problem now. Maybe it's just me then.

 

[busydude]

No, I haven't grabbed the F1 game yet but I will once price drops a bit more. For example, some websites got 40-50 fps on Radeon 4890 with 4AA at 1920x1080 in Dirt 2 w/DX9. However, I couldn't play that game with my card on those settings even after overclocking it. So to me it was unplayable and choppy with 4AA despite averaging 45 fps in my own rig in the built-in benchmark. I had to reduce to 2AA or 0AA even. With 470 I don't have a problem now. Maybe it's just me then.

F1 is really smooth even at those really low FPS, when I fired up the game I thought I was doing over 60, fired up FRAPS and I was shocked that I was actually doing around 32-35 FPS.. with the new patch and crossfire profiles I am currently doing ~45 Avg.. everything max at 1080P.

 

[Skyjoo]

I played F1 2010, Ferrari Virtual Academy, and the Williams FW31 in Iracing. Like people have said, the fluctuation of fps is the biggest problem when you're driving cars that are capable of huge acceleration/deceleration/cornering.

 

[RussianSensation]

F1 is really smooth even at those really low FPS, when I fired up the game I thought I was doing over 60, fired up FRAPS and I was shocked that I was actually doing around 32-35 FPS.. with the new patch and crossfire profiles I am currently doing ~45 Avg.. everything max at 1080P.

I was checking out youtube videos of F1. A lot of people noted the game runs at 45-60 fps without FRAPs, but drops to 35-40 fps with FRAPs recording. Strange outcome. It seems you have ran into that too.

 

[notty22]

I agree alot with needing higher fps in racing games. Its mostly why I upgraded. I now have everything maxed out in NFS Shift, even 15 cars racing right in front of me.
Just upgraded F1 with the dx11 patch to , all cranked up, the built in benchmark is longer than dirt2's. I use MSI AB, and leave the OSD on all the time. In this game have not checked, but in most, its only a 1 or 2% hit on fps.
Forza, by the way, is xbox 360 only right ? I used to think that was pretty good on xbox 360, demo

Not sure if the patch helped, but I tweaked some settings in the advanced wheel options for the xbox controller in F1 and now the racing is MUCH better. It felt like I could not make a turn, it was 'tight' no matter what. I have to say, it has a much more F1 feel to it now. Sound, acceleration, and finally steering.

 

[Keysplayr]

I usually like buying both top end cards and compare them myself.

 

[apoppin]

No, I haven't grabbed the F1 game yet but I will once price drops a bit more. For example, some websites got 40-50 fps on Radeon 4890 with 4AA at 1920x1080 in Dirt 2 w/DX9. However, I couldn't play that game with my card on those settings even after overclocking it. So to me it was unplayable and choppy with 4AA despite averaging 45 fps in my own rig in the built-in benchmark. I had to reduce to 2AA or 0AA even. With 470 I don't have a problem now. Maybe it's just me then.

Is $26.79 a sale for you?

http://store.steampowered.com/app/44310/
^_^

i got F1 2020 and i will be using the (new) built-in bench for my testing (if i like the game); unfortunately i have a severe D/L cap imposed by my Satellite ISP and it will take me awhile to actually get it.

I usually like buying both top end cards and compare them myself.

Which "both" are you talking about? When is the last time you bought any GTX?
()
.. for me, i like getting both top end cards and evaluating them myself; i haven't actually "bought" a graphics card in 2-1/2 years; my "collection" is up to 17 now

 

[Skurge]

Which "both" are you talking about? When is the last time you bought any GTX?
()

If I remember correctly, he bought an XFX 5870 and compared it to a 480. but I see what you did there.

 

[soccerballtux]

Idk I'm pretty happy with anandtech. I've ventured elsewhere but never really found a reliable source. Xbitlabs is nice in general IMO, but I haven't watched them closely.

 

[soccerballtux]

i find it very difficult to get a thirty page review out - TWELVE THOUSAND WORDS; using 25 benchmarks with 6 card configurations at two or three resolutions - all with FRESH benchmarks - in 7 days from receipt of card to publication when NDA ended

The OP and the rest of you need to realize that i spent 20 hours a day for 7 days working on it. Do you guys work this hard at your own jobs?

Did you want more?
- i'm working on cloning myself
:biggrin:

No, and No.

We weren't meant to work this hard. It's one of the reasons I'm becoming anti-capitalism-- because pursuit of greed has forced many of us to sell our lives just to make a moderate living or excel in our work. Jesus never said we should have to work 20 hours/day/7 to make a living. These things are a result of greed.

 

[soccerballtux]

I disagree with that. Hard OCP's videocard reviews are too subjective. For example, I am not happy playing a racing game below 60fps, but I am perfectly fine at 35 fps in Crysis. Kyle might be OK with 45 fps in Dirt 2 but may want 60 fps in Crysis.

I want average and minimum framerates at the same settings and I will decide myself what I consider "playable". Because he tests videocards at completely different settings to arrive at what he considers playable (not what I consider playable), his tests have questionable value. What is the point of testing high end cards only at 2560x1600 4AA in Metro, to get 25-30 fps? That's not how I am going to play the game. I am going to run at with Tessellation at 1680x1050 or with noAA/no tessellation at 1920x1080 and HardOCP is not going to provide with me either of these scenarios!! HardOCP is getting way too lazy. They never test enough games.

They got a 1.7 fps disadvantage with HD5870 compared to GTX480 at 2560x1600 with Tessellation ON in Metro 2033 and only 2.2 fps disdvantage in Civilization 5. Those results are completely off. It takes 5 min to check TechSpot, LegionHardware, Xbitlabs, Tom's Hardware, Tech Report, etc. to see that GTX480 is far faster than HD5870 in both of those games.

Yeah I've been thoroughly unimpressed with HardOCP's reviews. I saw a few where they were completely unfair to AMD, and never looked back.

 

[soccerballtux]

i find it very difficult to get a thirty page review out - TWELVE THOUSAND WORDS; using 25 benchmarks with 6 card configurations at two or three resolutions - all with FRESH benchmarks - in 7 days from receipt of card to publication when NDA ended

The OP and the rest of you need to realize that i spent 20 hours a day for 7 days working on it. Do you guys work this hard at your own jobs?

Did you want more?
- i'm working on cloning myself
:biggrin:

What if certain reviewers chose to take a certain section of performance and excel at it? If sites only covered perhaps one thing in depth, like SLI, we could have much more thorough reviews all around. If everyone stuck to their guns, I don't think readership would suffer, because we would still be forced to visit all the sites we were visiting before if we wanted a complete picture.

 

[MangoX]

I too have been feeling this for a while now. I like [H]'s review line graphs as they show framerates over a time sample. Good to know if the min/average framerate is consistent or not. It would be nice if we scrape off the bottom 5% from the minimum fps or just average out the minimums. I wonder what the numbers will look like with that. I suggest this because with normal graphs a min fps of 15 doesn't tell us if the framerate is consistently dropping to the mid 10's or if that's just one anomaly where the FPS dipped because of network/hdd lag or other.

 

[soccerballtux]

Is $26.79 a sale for you?

http://store.steampowered.com/app/44310/
^_^

i got F1 2020 and i will be using the (new) built-in bench for my testing (if i like the game); unfortunately i have a severe D/L cap imposed by my Satellite ISP and it will take me awhile to actually get it.

Which "both" are you talking about? When is the last time you bought any GTX?
()
.. for me, i like getting both top end cards and evaluating them myself; i haven't actually "bought" a graphics card in 2-1/2 years; my "collection" is up to 17 now

lolsatelliteinterent

 

[soccerballtux]

I too have been feeling this for a while now. I like [H]'s review line graphs as they show framerates over a time sample. Good to know if the min/average framerate is consistent or not. It would be nice if we scrape off the bottom 5% from the minimum fps or just average out the minimums. I wonder what the numbers will look like with that. I suggest this because with normal graphs a min fps of 15 doesn't tell us if the framerate is consistently dropping to the mid 10's or if that's just one anomaly where the FPS dipped because of network/hdd lag or other.

if fps is consistently dropping then it'll show up in the average.
I see your point though. I've always wondered that too-- sometimes you get something with a minfps of like 8 and it's like, really? Does it do that for a long time? Or is that just a fluke and wouldn't be there if you retested? I think another poster mentioned HardOCPs graphs in relation to this...that would be one thing they are good at I guess. AMD usually fares worse than Nvidia it seems in those though and that makes me unhappy

 

[taltamir]

@OP, your demands are far too high. It takes time and money to do what you say. I am satisfied with sites that have good methodology like H, and excellent in depth analysis of new tech like Anand... I am ecstatic for tools like anand's bench...
it would be awesome if every review included dozens of cards all retested with latest drivers / game versions as of the review's launch... but it is completely unrealistic to make such demands, nobody has the time and money to do that.

 

[ModestGamer]

I disagree with that. Hard OCP's videocard reviews are too subjective. For example, I am not happy playing a racing game below 60fps, but I am perfectly fine at 35 fps in Crysis. Kyle might be OK with 45 fps in Dirt 2 but may want 60 fps in Crysis.

I want average and minimum framerates at the same settings and I will decide myself what I consider "playable". Because he tests videocards at completely different settings to arrive at what he considers playable (not what I consider playable), his tests have questionable value. What is the point of testing high end cards only at 2560x1600 4AA in Metro, to get 25-30 fps? That's not how I am going to play the game. I am going to run at with Tessellation at 1680x1050 or with noAA/no tessellation at 1920x1080 and HardOCP is not going to provide with me either of these scenarios!! HardOCP is getting way too lazy. They never test enough games.

They got a 1.7 fps disadvantage with HD5870 compared to GTX480 at 2560x1600 with Tessellation ON in Metro 2033 and only 2.2 fps disdvantage in Civilization 5. Those results are completely off. It takes 5 min to check TechSpot, LegionHardware, Xbitlabs, Tom's Hardware, Tech Report, etc. to see that GTX480 is far faster than HD5870 in both of those games.

which cards exactly say at $200 and up can't do that ? Seriously ? Then again are you a hardcore racer with a logictech g27 wheel or a ECS like myslef or are you a mouse and keyboard warrior ?

Becuase none of those games are taxing to even a 56xx card or a gtx450 aside from dirt2 which has lots of wonderful eyecandy but severly lacks in many ways on gameplay and mods.

Rfactor can get 60fps on most onboard video and its a pretty serious gaming community.

I mean if your wanting to talk about arcade style racing games vrs racing simulations thats fine but even the low end hardware is pretty damn capable of "Real Race simulations"

What exactly are you complaining about ? You are always complaining about something. Its like if the results weren't what you expected you complain about the test, if you want nvidia cards to look the best in benchs, just run nvidia sponsored games. That should solve the problem.

 

[SirPauly]

Considering the time they may have there is so much they can really test and probably doing the best they can. The key is with owners of the hardware, there is much more testing and evaluation, so many different subjective tastes -- that is why owners may differ from reviewers, too.

Also, reviewers are individuals with their own subjective tastes and may evaluate things differently than certain owners.

I doubt it is conspiracy or reviewers getting payed off and what I do is just read as many different reviews as I can considering one reviewer may test differently or may have different subjective views or a different data point that may be important.

For me texture quality is paramount and flexibility of tools is essential but to a reviewer one may deem something "good enough" and perfectly happy with application AA and AF.

One of the problems I did have was with sound and thermals with many reviewers simply testing in open environments and yet not testing in closed environments like what many consumers may actually have. It was hard to find data in a closed environment with good air flow like I game with.

 

[ModestGamer]

a 5770 can run nfs shift at stock clocks around 60fps and 905 of the other racing games at 1920x1080 on high and extreme detail settings. The extreme settings are a bit of a waste though becuase its wasted detail you normally can't see anyways.

BTW get a real racing controller.

g25/g27 thrustmaster, ecs etc.

trying to drive a race cars with a controller is like trying to balance the federal budget, both are futile efforts and even if you pull either off you'll never feel right about doing it.

http://www.hardocp.com/article/2009/10/12/amd_ati_radeon_hd_5770_5750_review/5

http://www.overclock.net/ati/664273-hd5770-dirt-2-benchmark.html

FPS won't make up for a bad controller strategy.

I agree alot with needing higher fps in racing games. Its mostly why I upgraded. I now have everything maxed out in NFS Shift, even 15 cars racing right in front of me.
Just upgraded F1 with the dx11 patch to , all cranked up, the built in benchmark is longer than dirt2's. I use MSI AB, and leave the OSD on all the time. In this game have not checked, but in most, its only a 1 or 2% hit on fps.
Forza, by the way, is xbox 360 only right ? I used to think that was pretty good on xbox 360, demo

Not sure if the patch helped, but I tweaked some settings in the advanced wheel options for the xbox controller in F1 and now the racing is MUCH better. It felt like I could not make a turn, it was 'tight' no matter what. I have to say, it has a much more F1 feel to it now. Sound, acceleration, and finally steering.

 

[notty22]

a 5770 can run nfs shift at stock clocks around 60fps and 905 of the other racing games at 1920x1080 on high and extreme detail settings. The extreme settings are a bit of a waste though becuase its wasted detail you normally can't see anyways.

BTW get a real racing controller.

g25/g27 thrustmaster, ecs etc.

trying to drive a race cars with a controller is like trying to balance the federal budget, both are futile efforts and even if you pull either off you'll never feel right about doing it.

http://www.hardocp.com/article/2009/10/12/amd_ati_radeon_hd_5770_5750_review/5

http://www.overclock.net/ati/664273-hd5770-dirt-2-benchmark.html

FPS won't make up for a bad controller strategy.

Making excuses what others need for image quality settings, does not make the 5770 any shinier either.

 

[apoppin]

No, and No.

We weren't meant to work this hard. It's one of the reasons I'm becoming anti-capitalism-- because pursuit of greed has forced many of us to sell our lives just to make a moderate living or excel in our work. Jesus never said we should have to work 20 hours/day/7 to make a living. These things are a result of greed.

This is not how i make my living. i retired and this is what i do because i love to do it. The ONLY thing i don't like is being under a deadline where i have to spend 20 hours a day with the HW and writing. Fortunately so LITTLE of my year is in this kind of panic. Only recently has under-NDA HW been coming out at a fast clip.
-- right now i am writing an article that i benched back with Cat 10.6; it will never be out of date, however, as it deals with CPU scaling - so it got put on a back-burner.

What if certain reviewers chose to take a certain section of performance and excel at it? If sites only covered perhaps one thing in depth, like SLI, we could have much more thorough reviews all around. If everyone stuck to their guns, I don't think readership would suffer, because we would still be forced to visit all the sites we were visiting before if we wanted a complete picture.

i don't know about other reviewers, but that IS what i attempt to do. i focus on what i am particularly interested in - game performance and that is why i have 25 benchmarks; half older games and half new. i think to make a $500 decision, you should visit several review sites that specialize in what you are interested in besides the few favorites sites you may have.

lolsatelliteinterent

Super-overpriced archaic service. Hopefully they are expanding it this coming year. i won't move out of the rural area that i love just for an uncapped D/L. Besides, it gives me an excuse to visit my cousin and use his 10MB/s line to also d/l a few games that i got the month before
:sneaky:

 

[taltamir]

No, and No.

We weren't meant to work this hard. It's one of the reasons I'm becoming anti-capitalism-- because pursuit of greed has forced many of us to sell our lives just to make a moderate living or excel in our work. Jesus never said we should have to work 20 hours/day/7 to make a living. These things are a result of greed.

seriously? Do you have ANY idea how much hard work it is to be a hunter gatherer? you work every waking moment of your life and STILL might starve come winter...
And subsistence farming? not much better... you work every waking moment as well and can still starve, just less likely to starve and actually own a house that protects your from the weather.

Capitalism gave us the ability to go the store and buy enough food for a day at the cost of mere hours of work (a few hours even at minimum wage more than pays for a day's food; if you are buying cheap food like ramen... well, 2 minutes of work at minimum wage is enough to buy one)... it gives us tools like a dishwasher and clothes washer that take long arduous tasks and make them a breeze (so we can always have clean clothes instead of wallowing in filth and disease)... indoors showers and toilets, toilet paper, soap, shampoo (so we can be clean much less diseased)... And let us not forget the tools of modern agriculture that moves farmers from barely making enough food to survive to producing enough to feed 10 people per farmer.

And if you choose to be greedy than by definition nobody is forcing you to do anything, you are choosing it. Only in a capitalist society can you CHOOSE to work a minimum amount of hours to meet your minimal needs/desires, and then play with your computer the rest of the day... In communist Russia very bad things happened to you if you decided to just "skip work" one day

 

[ModestGamer]

Making excuses what others need for image quality settings, does not make the 5770 any shinier either.

thats not my point. My point is that people are saying they need 60fps which is BS and that you need a $500 card to do so which is also BS. IQ on a $200 card these days playing modern racing games is more then adequate and will often dilver over 60fps.

I think the real reason so many feel they need massive FPS is becuase they will spend $700 on a card but neglect to buy the proper devices with which to pilot there endevors.

I had no problem playing NFS on a onboard geforce 8300 at medium settings at 1440x900 and routinely did well in both the game and online challenges. Most of this was due to the fact that I have a good control solution which makes the vehicle control better.

I think its a false paradigm. Poor performance in racing games has little to do with IQ or FPs and more to do with ability and a good method to control the car.

But I digress

On any given night there might be 40,000 people racing nascar etc on rfactor " combined public and private servers.

same reason as to why I was able to consistnely score well on halo CE running at super slow frame rates of 30fps and 1024x768 resolutions.

Its not about the hardware. It usually about proficiency and people who chase hardware usually have low proficiency. the only people who might be hardware capped are very talented players or professionals where the hardware could be a limiting factor but more then likely it is not and they will still kickass regardless of what system they are on.

So take that for what ya will.