Sign in with
Sign up | Sign in
Your question

5770 or 5850?

Last response: in Graphics & Displays
Share
February 26, 2010 5:43:20 PM


I'm purchasing a new computer soon (dont prefer building my own) and I'm looking at 2 systems. One has a 5770 and the other has a 5850.

I plan on doing some gaming on either a 23" or 24" monitor. I'm not a hardcore gamer by any means and I dont game for hours at a time, but I would like to be able to buy the latest title and be able to run it well.

I feel like the 5850 might be overkill for my needs but I just a read a huge thread on another forum about "grey screen" issues with these cards. Dont know if it matters that I'm buying a completely new computer with one of these cards, but I would imagine they're using the same drivers.

Should I look for a computer with a 4890 instead? Or a comparable GeForce card? Not sure if I really need Directx11 right now.

Thoughts....?

More about : 5770 5850

February 26, 2010 5:47:35 PM

HD5850 is the best bang for the buck so far and you will definitely need it to have high game setting with 23"/24" monitor!

BTW, go for HD4890 if you are considering HD5770. This is just my opinion though, you are the one to decide.
m
0
l
a c 235 U Graphics card
February 26, 2010 5:48:31 PM

Directx11 cards will have a longer lifespan then Dx10 like the 4890

ATI fixed the grey line issue, just download the latest driver from their web site

if money is not an issue, get the 5850. It will work with monitor better with the latest games. The 5770 is a solid card and performs well but the 5850 is better.

m
0
l
Related resources
a b U Graphics card
February 26, 2010 6:06:35 PM

I would opt for a 4890 as well but since the OP has only two 5 series cards to choose from I would go with the 5850, specially with a 23/24" screen ..
m
0
l
February 26, 2010 6:17:37 PM

Can you list some specific titles you plan on playing? This can help determine what card would be best for you (not to mention possibly save you a bit of green).

For now, though, I say the 5770. It'll handle just about anything you throw at it at 1920x1080 resolutions with little to no problems at all, and isn't as expensive as the 5850.
The alternative would be the new 5830. Sure, it's performance is "on par" with the slightly cheaper 4890, but the 4xxx series is going the way of the dodo (corporate greed breeds technological progression), and we're only talking ~$30 difference. For that extra $30, you get future-proofing (via DX11 support), not to mention Eyefinity (will make you want to add a couple more 23" monitors). And it'll be cheaper than the 5850 while performing better than the 5770.
m
0
l
February 26, 2010 6:39:02 PM


Thanks for the input everyone.


Here's what I plan on playing:

Crysis Warhead
Diablo 3
Assassin's Creed 2
Dragon Age Origins
and maybe something like WOW or Star Trek Online..

m
0
l
February 26, 2010 6:57:23 PM

Yea, going by what you've got listed there, the 5770 should be able to power through those games on a 23" monitor with no problem. Of course, things like the CPU will also be a factor in some instances, but the 5770 will have no trouble.
Save the extra bit of scratch, but it towards buying Diablo 3 and the expansion for Dragon Age: Origins (The Awakening, I believe is what it's called).
m
0
l
a c 235 U Graphics card
February 26, 2010 7:23:40 PM

the 5850 for Crysis warhead

the 5770 can run the others games at high (not very/ultra high)

like I said 5850 if you can afford it, if you don't care about bells and whistles and want to spend the money elsewhere, the 5770 is still a solid card
m
0
l
February 26, 2010 8:54:43 PM


FYI - the cpu will be an i7-860.


Thanks again for all the helpful comments. I'm leaning more and more towards the 5770 at this point.
m
0
l
February 26, 2010 9:12:29 PM

The Core i7 will definitely help keep your frames up there. Crysis Warhead may require turning down a setting or two (at most) to get some consistently good frame rates, but overall, the 5770 will deliver quite a bit of eye candy for ya.

Enjoy your new card! :D 
m
0
l
a c 229 U Graphics card
February 26, 2010 9:34:52 PM

Monitor size doesn't tell as as much as resolution....The 24" normal resolution is 1920 x 1200 or 1920 x 1080.

The 48xx series recommended above will do 30 fps in most DX9/10 games, but does not support DX11.

The 57xx series recommended above will do 30 fps in most DX9/10 games, but not in Dx11

The 58xx series recommended above will do 30 fps in most DX9/10/11 games
m
0
l
a c 235 U Graphics card
February 26, 2010 10:30:37 PM

JackNaylorPE said:
Monitor size doesn't tell as as much as resolution....The 24" normal resolution is 1920 x 1200 or 1920 x 1080.

The 48xx series recommended above will do 30 fps in most DX9/10 games, but does not support DX11.

The 57xx series recommended above will do 30 fps in most DX9/10 games, but not in Dx11

The 58xx series recommended above will do 30 fps in most DX9/10/11 games


5770 1920x 1200 - ultra high - 4x AA

30 min - 34 max FPS in DX11

http://www.tomshardware.com/reviews/radeon-hd-5830,2564...

guess you are wrong!
m
0
l
a c 229 U Graphics card
March 1, 2010 5:41:11 PM

ct1615 said:
5770 1920x 1200 - ultra high - 4x AA

30 min - 34 max FPS in DX11

http://www.tomshardware.com/reviews/radeon-hd-5830,2564...

guess you are wrong!


Nope, I'm just better at math.... 29.5 is < 30

Better go back to your link and read again.

I''m reading Radeon HD 5770.

33.5 fps max (meaningless statistic)
29.5 min

It's not about hitting a max fps and some point in the game, it's about "maintaining" a minimum fps throughout the game. The 5770 can NOT maintain 30 fps....and this in a game in which the DX 11 implementation is so minimal that the reviewers could hardly tell when DX11 was enabled.
m
0
l
a c 235 U Graphics card
March 1, 2010 6:44:57 PM

JackNaylorPE said:
Nope, I'm just better at math.... 29.5 is < 30

Better go back to your link and read again.

I''m reading Radeon HD 5770.

33.5 fps max (meaningless statistic)
29.5 min

It's not about hitting a max fps and some point in the game, it's about "maintaining" a minimum fps throughout the game. The 5770 can NOT maintain 30 fps....and this in a game in which the DX 11 implementation is so minimal that the reviewers could hardly tell when DX11 was enabled.


once again you fail in your statement; 29.5 is 30fps now leave this forum and leave us of your worthless stupidly

CT1615 - 2
JackNaylor - 0 (you should be familiar by now of being called a zero or failure in life)
m
0
l
a b U Graphics card
March 1, 2010 8:51:53 PM

"now leave this forum and leave us of your worthless stupidly

CT1615 - 2
JackNaylor - 0 (you should be familiar by now of being called a zero or failure in life)"

Please try not to offend others in your posts.

My opinion is to buy the 5850(system) if you want better FPS and more effects witch make the game more enjoyable. If no, the 5770(system) is ok too.
m
0
l
a c 229 U Graphics card
March 5, 2010 2:55:33 AM

quotemsg=2088569,15,329690]once again you fail in your statement; 29.5 is 30fps[/quotemsg]

No, 29.5 is 29.5

If 29.5 is 30, then I'd ask someone to lend me $1,000 every Monday and I'll pay them back $980 on every Friday....."because $29.50 is $30.00"
m
0
l
March 5, 2010 3:40:44 AM

Can you tell the difference between 29.5fps and 30fps?

No and that's why ct1615 said 29.5fps=30fps. Although he is rude by calling you stupid, I have to agree with him on 29.5fps=30fps.
m
0
l
a b U Graphics card
March 5, 2010 5:35:16 AM

JackNaylorPE said:
Nope, I'm just better at math.... 29.5 is < 30

Better go back to your link and read again.


WTF !?!

You're thinking that the 29.5 defence is good enough after your own statement? I think you should re-read your own comment again. :heink: 

Fine if that 29.5 is too low, then turn off the 4X AA, and guess what it'll go above your 29.5 fps as a MINIMUM fps which isn't even what you said, you simply said it won't do 30fps, so if you want to be that anal Jack, then even by your own math, YOU FAIL ! :lol: 

St00pid argument, since once again it's proven the HD5770 does better than you give it credit for. :sarcastic: 

Once again you make up these false limits for the HD5770, and at every turn you get proven wrong. [:thegreatgrapeape:6]
m
0
l
a c 229 U Graphics card
March 6, 2010 6:42:57 PM

Ape:

I stated that it scored < 30 .....the statement is wholly and truly correct. Ct is the one who took issue with it and said "you are wrong".

My statement was that the 5770 can't "do 30 fps". If it necessary to qualify every single time that when we are talking about "doing 30 fps" we are talking providing a stutter free game experience by "maintaining a minimum of 30 fps throughout the game w/o turning down detail or disabling AA or other features" ...then mea culpa on my part for believing that a "generally accepted industry standard" would "go without saying" among enthusiasts. Most of the people buying GFX cards are not interested in that a card can do 90fps at certain points in the game or that averaging those high points with the low points gets an average above some number.....they are concerned about:

Will the game lag ?
Will it stutter ?

30 fps is the defacto accepted industry standard as a minimum for determination of acceptable performance. Can I claim to be a member of the 4 Ghz OC club with a BCLK of 199 and a CPU multiplier of 20 on a 920 ? Are you going to tell me that if you play a game that does 40 fps half the time and 20 fps half the time, you will be content with the experience ? If a mountain road has a minimum speed limit of 40 and you can only do 30 up hill, can you argue the ticket you get in court by claiming that downhill you go 50 mph so it averages out ? I don't think the judge is gonna be OK w/ the fact that you are only endangering people half the time.

The speed limit is number that is not to be crossed. 30 fps is the generally accepted minimum limit that a game can be played w/o diminishing the gaming experience. Reducing detail settings to get to that point doesn't count. The fact is the 5770 can not maintain a minimum 30 fps in Dirt2 .... It can't even maintain an average of 30, scoring an average of only 24 in the HH test. If someone wants to take issue with that, then they have to show me 30.0 or better, and not with detail settings turned off like in the THG Dirt2 review. I wouldn't argue that a user can distinguish 29.5 from 30. But I gotta take issue when someone starts saying it can do 30 fps (29.5 actually) and refuses to acknowledge min / avg scores of 15/24 on another test and doesn't mention that AA has to be tuned off in order to get that 29.5 score.

We see in the HardwareHeaven test, the Dirt2 minimum fps of 15 is at only 50% of the generally accepted standard, certainly something to be very concerned about. You can't only pick the tests that support your argument and ignore the ones that don't. The 15 fps is certainly distinguishable from 30 and again with no perceptible level of increased detail over DX9. Why the big difference ? I couldn't be sure at 1st but I have a question for the author of the article ct referenced.....seems his graph has an error. Looking up the test configuration in this article it says:

http://www.tomshardware.com/reviews/radeon-hd-5830,2564...

DiRT 2 Version 1.0.0, Custom THG Benchmark
Run 1: Ultra High Settings, No AA, DirectX 9
Run 2: Ultra High Settings, No AA, DirectX 11

So test config says no AA, even tho graph says AA was enabled in same article ....hmmm....typo perhaps ? ..... let's look up the original THG test that I have repeatedly referenced....what do we see.....

http://www.tomshardware.com/reviews/dirt-2-performance-...

We see that the 29.6 and 33.4 fps results were obtained with AA turned off......pretty important little piece of information to leave out of one's claim don't ya think. That certainly would explain the difference between 15 and 29.5 fps.

Those wanting to make believe that the 5770 will deliver a stutter free experience (w/o reducing detail / AA) in DX11 games of today and the next 24 months will just have to get over it and recognize the numbers. Dirt2 simply can not deliver a minimum of 30 fps, nor get close to it without chopping back detail settings. We have one test where it apparently gets 29.5 but it seems that they had to turn AA off to get that score. We have another where AA is clearly on at 4AA and it scores 15 fps. The 15 was obtained on a test system w/ an Intel 975 OC'd to 4ghz and memory from @ 1800mhz so the hardware was certainly no bottleneck.

http://www.hardwareheaven.com/reviews.php?reviewid=902&...

Your standing on shaky ground if you wanna say my "can't do 30 fps" statement is wrong when one test is 15 fps and the one that gets close to 30 w/ a 29.5 doesn't bother to mention that they had to turn off AA settings to get that close to 30.

And before the "it's only one game" argument is brought to bear yet again. I have two answers:

1. It can't maintain the generally accepted minimum of 30 fps in AvP either.

http://www.hardwareheaven.com/reviews.php?reviewid=902&...
http://www.hardwareheaven.com/reviews.php?reviewid=937&...

On DX11 titles.....@ 1920 x 1200 resolution

AvP (Min / Avg / Max fps) 2AA

5770 (21/30/43) .... are we going to argue that 21 is close enough to 30 ?
5830 (30/43/63)
5850 (36/49/70)

Dirt2 (Min / Avg / Max fps) 4AA

5770 (15/24/54) .... are we going to argue that 15 (or how about the avg fps of 24) is close enough to 30 ?
5850 (31/42/70)
5870 (42/52/85)

2. The fact that THG also said that you can barely distinguish the Dirt2 game experience from DX9 to DX11, that this minimal difference had a "colossal" impact on performance and that they apparently had to turn off AA to get even those scores is damning. What is going to happen with the DX11 games of XMas 2010 and Xmas 2011 which will certainly have more extensive incorporation of Dx11 technology ? If it can't deliver DX11 above 30 fps with a game which has only minimal incorporation of Dx11 features, what's it going to be doing 18 months from now when devs have had almost 2 years to implement the technology to its fullest extent ? Using the 5770 as a DX11 card for 1920 x 1200 resolution will unavoidably result in a diminished game experience; it can't maintain 30 fps in AvP or Dirt2. It can't even average above 24 fps with 4AA on. If you want to assure a stutter free experience @ 1920 x 1200 w/ turning off detail (AA), the 58xx series cards can provide that, the 57xx cards can't.

In summary:

In Dirt2, under DX11 w/ 1920 x 1200 resolution and no AA, minimum fps is 29.5
In AvP, under DX11 w/ 1920 x 1200 resolution and 2 AA, minimum fps is 21
In Dirt2, under DX11 w/ 1920 x 1200 resolution and 4 AA, minimum fps is 15

Looks like a pretty even progression from 15 to 29.5 ....I for one can not recommend a card at a given resolution whereby the only way it can maintain generally accepted minimum frame rates is if the user turns down detail settings. Telling people the card is "good" for 1920 x 1200 w/o mentioning the fact that you have to play at reduced settings is not the kind of advice people come here looking for.



m
0
l
March 6, 2010 6:58:08 PM

JackNaylorPE said:
Ape:

I stated that it scored < 30 .....the statement is wholly and truly correct. Ct is the one who took issue with it and said "you are wrong".

My statement was that the 5770 can't "do 30 fps". If it necessary to qualify every single time that when we are talking about "doing 30 fps" we are talking providing a stutter free game experience by "maintaining a minimum of 30 fps throughout the game w/o turning down detail or disabling AA or other features" ...then mea culpa on my part for believing that a "generally accepted industry standard" would "go without saying" among enthusiasts. Most of the people buying GFX cards are not interested in that a card can do 90fps at certain points in the game or that averaging those high points with the low points gets an average above some number.....they are concerned about:

Will the game lag ?
Will it stutter ?

30 fps is the defacto accepted industry standard as a minimum for determination of acceptable performance. Can I claim to be a member of the 4 Ghz OC club with a BCLK of 199 and a CPU multiplier of 20 on a 920 ? Are you going to tell me that if you play a game that does 40 fps half the time and 20 fps half the time, you will be content with the experience ? If a mountain road has a minimum speed limit of 40 and you can only do 30 up hill, can you argue the ticket you get in court by claiming that downhill you go 50 mph so it averages out ? I don't think the judge is gonna be OK w/ the fact that you are only endangering people half the time.

The speed limit is number that is not to be crossed. 30 fps is the generally accepted minimum limit that a game can be played w/o diminishing the gaming experience. Reducing detail settings to get to that point doesn't count. The fact is the 5770 can not maintain a minimum 30 fps in Dirt2 .... It can't even maintain an average of 30, scoring an average of only 24 in the HH test. If someone wants to take issue with that, then they have to show me 30.0 or better, and not with detail settings turned off like in the THG Dirt2 review. I wouldn't argue that a user can distinguish 29.5 from 30. But I gotta take issue when someone starts saying it can do 30 fps (29.5 actually) and refuses to acknowledge min / avg scores of 15/24 on another test and doesn't mention that AA has to be tuned off in order to get that 29.5 score.

We see in the HardwareHeaven test, the Dirt2 minimum fps of 15 is at only 50% of the generally accepted standard, certainly something to be very concerned about. You can't only pick the tests that support your argument and ignore the ones that don't. The 15 fps is certainly distinguishable from 30 and again with no perceptible level of increased detail over DX9. Why the big difference ? I couldn't be sure at 1st but I have a question for the author of the article ct referenced.....seems his graph has an error. Looking up the test configuration in this article it says:

http://www.tomshardware.com/reviews/radeon-hd-5830,2564...

DiRT 2 Version 1.0.0, Custom THG Benchmark
Run 1: Ultra High Settings, No AA, DirectX 9
Run 2: Ultra High Settings, No AA, DirectX 11

So test config says no AA, even tho graph says AA was enabled in same article ....hmmm....typo perhaps ? ..... let's look up the original THG test that I have repeatedly referenced....what do we see.....

http://www.tomshardware.com/reviews/dirt-2-performance-...

We see that the 29.6 and 33.4 fps results were obtained with AA turned off......pretty important little piece of information to leave out of one's claim don't ya think. That certainly would explain the difference between 15 and 29.5 fps.

Those wanting to make believe that the 5770 will deliver a stutter free experience (w/o reducing detail / AA) in DX11 games of today and the next 24 months will just have to get over it and recognize the numbers. Dirt2 simply can not deliver a minimum of 30 fps, nor get close to it without chopping back detail settings. We have one test where it apparently gets 29.5 but it seems that they had to turn AA off to get that score. We have another where AA is clearly on at 4AA and it scores 15 fps. The 15 was obtained on a test system w/ an Intel 975 OC'd to 4ghz and memory from @ 1800mhz so the hardware was certainly no bottleneck.

http://www.hardwareheaven.com/reviews.php?reviewid=902&...

Your standing on shaky ground if you wanna say my "can't do 30 fps" statement is wrong when one test is 15 fps and the one that gets close to 30 w/ a 29.5 doesn't bother to mention that they had to turn off AA settings to get that close to 30.

And before the "it's only one game" argument is brought to bear yet again. I have two answers:

1. It can't maintain the generally accepted minimum of 30 fps in AvP either.

http://www.hardwareheaven.com/reviews.php?reviewid=902&...
http://www.hardwareheaven.com/reviews.php?reviewid=937&...

On DX11 titles.....@ 1920 x 1200 resolution

AvP (Min / Avg / Max fps) 2AA

5770 (21/30/43) .... are we going to argue that 21 is close enough to 30 ?
5830 (30/43/63)
5850 (36/49/70)

Dirt2 (Min / Avg / Max fps) 4AA

5770 (15/24/54) .... are we going to argue that 15 (or how about the avg fps of 24) is close enough to 30 ?
5850 (31/42/70)
5870 (42/52/85)

2. The fact that THG also said that you can barely distinguish the Dirt2 game experience from DX9 to DX11, that this minimal difference had a "colossal" impact on performance and that they apparently had to turn off AA to get even those scores is damning. What is going to happen with the DX11 games of XMas 2010 and Xmas 2011 which will certainly have more extensive incorporation of Dx11 technology ? If it can't deliver DX11 above 30 fps with a game which has only minimal incorporation of Dx11 features, what's it going to be doing 18 months from now when devs have had almost 2 years to implement the technology to its fullest extent ? Using the 5770 as a DX11 card for 1920 x 1200 resolution will unavoidably result in a diminished game experience; it can't maintain 30 fps in AvP or Dirt2. It can't even average above 24 fps with 4AA on. If you want to assure a stutter free experience @ 1920 x 1200 w/ turning off detail (AA), the 58xx series cards can provide that, the 57xx cards can't.

In summary:

In Dirt2, under DX11 w/ 1920 x 1200 resolution and no AA, minimum fps is 29.5
In AvP, under DX11 w/ 1920 x 1200 resolution and 2 AA, minimum fps is 21
In Dirt2, under DX11 w/ 1920 x 1200 resolution and 4 AA, minimum fps is 15

Looks like a pretty even progression from 15 to 29.5 ....I for one can not recommend a card at a given resolution whereby the only way it can maintain generally accepted minimum frame rates is if the user turns down detail settings. Telling people the card is "good" for 1920 x 1200 w/o mentioning the fact that you have to play at reduced settings is not the kind of advice people come here looking for.


how can you call that stupid lol...
m
0
l
a b U Graphics card
March 6, 2010 7:16:38 PM

JackNaylorPE said:
Ape:

I stated that it scored < 30 .....the statement is wholly and truly correct. Ct is the one who took issue with it and said "you are wrong".

My statement was that the 5770 can't "do 30 fps". If it necessary to qualify every single time that when we are talking about "doing 30 fps" we are talking providing a stutter free game experience by "maintaining a minimum of 30 fps throughout the game w/o turning down detail or disabling AA or other features" ...then mea culpa on my part for believing that a "generally accepted industry standard" would "go without saying" among enthusiasts.


That not a mea culpa, they involve contrition for the generalization and working towards something that matches the truth, not attacking the correct reasonig that showed your statement to be wrong. You were wrong, regardless of whether in your mind your reasoning was correct (which it still isn't, since you're wrong about the target fps since it's not exactly 30 either).

You specifically said;
"The 57xx series recommended above will do 30 fps in most DX9/10 games, but not in Dx11"

Not that t would score under <30. So not wholly and truly correct. You're first statement is a limitation (actually implying avg or max, not minimum), and your second one is a re-imagining to cover your previous mis-statement.

And talking about giving leeway for commonly held beliefs like 'industry standards', then it's funny you are so anal about the 29.5 =/= 30 fps, just like there is no set hardline of 30fps unlike you attempt to pretend that there is one. FPS require different level than RTS and RPGs and they also depend on complexity. And as far as 'hardcore gamers go, how diffifcult do you think it is doe them to overclock the core and memory 50hz to get you your 0.5 fps difference? Therefore it CAN even at 4XAA, it just didn't at default speed, which no 'hardcore gamer' would play at anyways. [:jaydeejohn]

Also, hardset 30fps is not based on experience in the game, it's based on the NTSC standard, which isn't 30hz according to you, since 29.97hz =/= 30hz.

Once again you make up false limitations, and also make errors while writing these limitations, and when proven wrong then you fall back on 29.5 fps =/= 30, when in fact that's not even what you said at all, thus moving your own goal-posts.

Perhaps think next time before you write, and make sure you write it to such an anal standard that even you taking the other position wouldn't rip it apart with something simpler than rounding error defense in an easily obtainable benchmark, just like the easily attainable ones showing the driver improvements your 'long search' failed to produce. :pfff: 
m
0
l
a c 229 U Graphics card
March 6, 2010 7:20:40 PM

Quote:
Fine if that 29.5 is too low, then turn off the 4X AA, and guess what it'll go above your 29.5 fps as a MINIMUM fps .....hen even by your own math, YOU FAIL ! :lol: 


Question......how do you turn it off when it is already turned off ?

http://www.tomshardware.com/reviews/dirt-2-performance-...

Who's laughing now ?....Seems like my math is still good..... Do you ever bother to check your facts before posting ?
Quote:

Once again you make up these false limits for the HD5770, and at every turn you get proven wrong..... once again it's proven the HD5770 does better than you give it credit for


False ? What's been proven other than you don't check your facts before posting ? If you want to counter, unsupported statements won't do it. Where's your published references ? The tests below cover 60% of the DX11 games out there right now. None of them show the 5770 maintaining a minimum of 30 fps w/o choking back on settings.....even w/ AA turned off (yeah let me throw my money at that so I can play w/ reduced detail), best it could do was 29.5.

SHOW ME something that says the 5770 can maintain a minimum of 30 fps w/o choking down AA.

http://www.hardwareheaven.com/reviews.php?reviewid=902&...
http://www.hardwareheaven.com/reviews.php?reviewid=937&...
http://www.hardwareheaven.com/articles.php?articleid=14...

On DX11 titles.....@ 1920 x 1200 resolution

AvP (Min / Avg / Max fps) 2AA

5770 (21/30/43) Can not maintain 30 fps (21 < 30)
5830 (30/43/63)
5850 (36/49/70)

Dirt2 (Min / Avg / Max fps) 4AA

5770 (15/24/54)
5850 (31/42/70) Can not maintain 30 fps (15 < 30)
5870 (42/52/85)

Battleforge (Min / Avg / Max fps) 4AA

5870 (30/43/62)

Well if the 5870, does 30 as a minimum, then the 5770 certainly can't.....using the ratios from Dirt2, it would calculate out to

5770 (11/20/39)


http://www.tomshardware.com/reviews/dirt-2-performance-...

Dirt2 (Min / Avg / Max fps) AA OFF

5770 (29.6/33.4/??)

http://www.tomshardware.com/reviews/radeon-hd-5830,2564...

DiRT 2 Version 1.0.0, Custom THG Benchmark
Run 1: Ultra High Settings, No AA, DirectX 9
Run 2: Ultra High Settings, No AA, DirectX 11
m
0
l
a b U Graphics card
March 6, 2010 7:39:13 PM

JackNaylorPE said:
Quote:
Fine if that 29.5 is too low, then turn off the 4X AA, and guess what it'll go above your 29.5 fps as a MINIMUM fps


Uhmmm......how do you turn it off when it is alreday turned off.

then even by your own math, YOU FAIL ! :lol: 


Once again Jack, like pointed out in the other thread to you about 5 times, different drivers. You have to rely on benchmarks that are no longer relevant or accurate in order to try and make your case. Face it, you're out of date Jack, get some new material. :kaola: 

The drivers are 9.12 in your link, 10.2 in the one we're directing you to, and as also pointed out to you one of the biggest boosts of those 10.2 drivers was in DiRT2, over 10% (closer to 20%) with and without AA, which we already pointed out to you, but here it is again to redresh you memory and show everyone else as well;



Quote:
Seems like my math is still good.


Your math may be good, but like I said in the other thread, your research and comprehension skills are terrible! [:thegreatgrapeape:2]

Quote:
Do you ever bother to check your facts before posting ?


Yes, but it's pretty damn obvious that YOU DON'T! Once again, holding you to your own standards Jack, YOU FAIL !! [:jaydeejohn]

[:thegreatgrapeape]
m
0
l
a c 229 U Graphics card
March 6, 2010 7:40:38 PM

TheGreatGrapeApe said:
Perhaps think next time before you write, and make sure you write it to such an anal standard that even you taking the other position wouldn't rip it apart with something simpler than rounding error defense in an easily obtainable benchmark, just like the easily attainable ones showing the driver improvements your 'long search' failed to produce. :pfff: 



It's not about 29.5 vs 30 , it's about 15 versus 30. Getting to 29.5 only when ya turn off AA doesn't count.

Again, lots of words, still no proof. SHOW ME. Where's ya published references ?

I have AvP, Battleforge and Dirt 2 on the table all showing that the 5770 can't do 30 fps w/o choking down on AA. You want to counter you have to SHOW ME benchies showing a 5770 doing a minimum of 30 fps in those games w/o choking down AA.

m
0
l
a c 229 U Graphics card
March 6, 2010 7:42:03 PM

TheGreatGrapeApe said:
Once again Jack, like pointed out in the other thread to you about 5 times, different drivers. You have to rely on benchmarks that are no longer relevant or accurate in order to try and make your case. Face it, you're out of date Jack, get some new material. [:thegreatgrapeape]


Still no link, may I read the article please ?

It appears now, with new drivers, that the 5770 can hit 30 fps in 1 of 3 DX11 games....still not impressed. That leaves us with 2 outta 3 with no evidence that it can.

Quote:
Your math may be good, but like I said in the other thread, your research and comprehension skills are terrible! . Yes, but it's pretty damn obvious that YOU DON'T! Once again, holding you to your own standards Jack, YOU FAIL !!



How so ? you referenced a THG article claiming it was 29.5 w/ AA on and it was actually w/ AA Off ...this refects on me somehow ?
m
0
l
a b U Graphics card
March 6, 2010 7:54:30 PM

JackNaylorPE said:
Still no link, may I read the article please ?

It appears now that the 5770 can hit 30 fps in 1 of 3 DX11 games....still not impressed.


I linked it to you in the other thread, where I told you it took 30seconds to find using google at work, here let me google that for you (Catalyst 10.2 + Dirt2);
http://lmgtfy.com/?q=catalyst+10.2+dirt2



but since your research skills are so terrible, here they are again first the one above, and then LegitReviews' as well;

http://www.fudzilla.com/content/view/17754/40/

http://www.legitreviews.com/article/1216/6/


m
0
l
a b U Graphics card
March 6, 2010 8:08:13 PM

JackNaylorPE said:
How so ? you referenced a THG article claiming it was 29.5 w/ AA on and it was actually w/ AA Off ...this refects on me somehow ?


Is your reading comprehension or even visual skills that Bad?

You have a habit of cutting & pasting THG links/articles without understanding what they contain, and now it's obvious you don't even really read them. [:thegreatgrapeape:5]

You may have something bookmarked you think is mine, but the two are not the same test they are from different reviews by Don even the endings to the truncated links should be obvious enough to you despite your impairments;

My and CT1615's link with 4XAA;
http://www.tomshardware.com/reviews/radeon-hd-5830,2564...

Your link with NO AA;
http://www.tomshardware.com/reviews/dirt-2-performance-...


Or to help you with pictures;



Different Benchmarks.

And considering you went through all the effort and page space above of linking to the configuration page to point out the DiRT2 settings, as if I couldn't figure that out: "Looking up the test configuration in this article it says: ", yet you completely missed the whole Catalyst version # in that same configuration page. :pt1cable: 

Wow you really do suck at researching your own statements. :lol: 
m
0
l
a c 235 U Graphics card
March 6, 2010 9:11:15 PM

JackNaylorPE said:
Ape:

30 fps is the defacto accepted industry standard as a minimum for determination of acceptable performance.


what industry? can you link it? who voted on this measure? So someone running Crysis at 28FPS is running the game at unacceptable performance? Will they not be able to enjoy the game as much as someone playing it at 31 FPS? Will the game simply uninstall itself before you even attempt to play it? Will the FBI come crashing through your front door and arrest you? Will your computer just explode like a star during a supernova?

People must be warned!!

DO NOT PLAY YOUR COMPUTER GAME AT UNDER 30 FPS!!!! IT IS BELOW INDUSTRY STANDARD* AND BAD THINGS WILL HAPPEN TO YOU!!!


* What specific industry is TBA.
m
0
l
a c 229 U Graphics card
March 6, 2010 9:14:29 PM



Fudzilla.....Now I understand the unwillingness to post that link....bastion of unbiased reporting that it's known to be, I ignore it on web searches. But I do thank you for the links as it now allows some intelligent discourse.

I took the effort to post my sources, don't see why you were so reluctant to provide the same courtesy. Looking at the new material, it seems apparent that the new drivers have done much to improve the 5770's performance ..... However, this test done after both your links were published shows only a 30 fps minimum w/ 2AA enabled:

http://www.hardwareheaven.com/reviews.php?reviewid=937&...

.....I think it's a given that at 4AA the result would be lower. Let me just say that I acknowledge the improvement with the new drivers but given the varied results, the jury's still out ..... don't have faith in fudzilla and the other two references have results on both sides of the 30 fps barrier at 4AA.

Now let's move along to AvP .... a game which just came out .... well after these new drivers were released and the review posted after both your links were published. Here the 5770 manages only 21 fps w/ 2AA.

Given the new information, I can move Dirt2 up into the "might do 30 w/ 4AA" since we have conflicting data in two different reviews. But that still leaves AvP, BF and STALKER on the table. So here's what we got for 1920 x 1200:

Aliens vs. Predator (21 min fps w/ 2AA)
BattleForge (30 min fps w/ 4AA on a 5870 .... rules out 5770 having a shot)
Battlefield: Bad Company 2 (No references found)
Dirt 2 (current tests on both sides of 30 fps w/ current drivers)
S.T.A.L.K.E.R.: Call of Pripyat (15 fps w/ old drivers at 4AA)

Most importantly, all of today's DX11 games have only minimal incorporation of DX11 technology. As THG called it, you could barely discern a difference between DX9 and 11 in Dirt2. What happens when devs have 18 - 24 months of development time under their belts and get Dx11 well integrated with their new releases ?

So where are we now ?

1. We have conflicting sources saying that the 5770 can / can not do 30 fps at 4AA. w/ latest drivers in Dirt2
2. We have 1 source saying that the 5770 can only do 21 fps in AvP w/ latest drivers
3. We have one source saying the 5870 can do 30 fps w/ 4AA w/ old drivers in BF..... take from that what you may about the 5770, but I'd say its a safe bet that it can't do what the 5770 can do.
4. We have 15 fps w/ old drivers and 4AA in STALKER and I think it's given driver optimizations haven't doubled that.

What can we conclude from this ?

-Well given the conflicting sources, I can't make a judgment on Dirt2....but let's give it a bye for arguments sake.
-It is safe to say that the 5770 can't do 30 at 4AA since in AvP only does 21 w/ 2AA as of 2 weeks ago's test.
-I'm thinking that the 5770 *might" climb to 30 fps in BF w/ latest drivers ... but can't say it's likely.
-I don't see new STALKER doubling it's performance w/ latest drivers.

While I wouldn't take it to the bank, I wouldn't bet against the 5770 doing 30 fps on 2 of the 4 DX11 games I have seen tested. That's not good enough for me to recommend this card to a person building a 1920 x 1200 box today. If it's barely making 30 w/ today's games having minimal incorporation of DX11, how will it stand up to the games of XMas 2011 ? Frankly, I think the 5870 and 480 will struggle w/ the most demanding 2 or 3 games coming out then but they will certainly fare better than the 5770 with most games.

In the end, I still can't recommend a card that, as far as I can determine, can't maintain 30 fps in at least 50% of today's DX11 games.



m
0
l
a b U Graphics card
March 6, 2010 10:25:30 PM

JackNaylorPE said:
Fudzilla.....Now I understand the unwillingness to post that link....bastion of unbiased reporting that it's known to be, I ignore it on web searches. But I do thank you for the links as it now allows some intelligent discourse.

I took the effort to post my sources, don't see why you were so reluctant to provide the same courtesy.


I had no unwillingness, the fact that I already showed you this in a previous thread not 4 days ago, and referenced that thread while posting that image in this one, means YOU already had been given these links as sources. :pt1cable: 
http://www.tomshardware.com/forum/forum2.php?config=tom...

My choice of the source is simple, since it was a previous driver specific bench at the specific resolution and game we were referencing, since you were being so specific. A bench of DX11 improvement that had already be brought to your attention previously. Nothing more, nothing less, and it's not like I distrust Fudzilla's benches as much as their rumours, these are tests supported by other sources which is why I provided the other link, since I knew when confronted with it you would use the Fudzilla cop-out. Even though if anything Fudzilla is notoriously anti-ATi rather than pro-ATi, so if anything one would expect them to underemphasis their improvement, not over-inflate them.

You can pretend I tried to hide something, but it's all there for everyone to see and always has been.

Quote:
Looking at the new material, it seems apparent that the new drivers have done much to improve the 5770's performance ..... However, this test done after both your links were published shows only a 30 fps minimum w/ 2AA enabled:

http://www.hardwareheaven.com/reviews.php?reviewid=937&...


Which still goes to the original point that it can in fact do 30fps, even by your own anal standards.

The rest is a much later addendum, especially for AvP which isn't even optimized yet, and irrelevant to whether it can do it or not.

Like every other time, the HD5770 is far better than you give it credit for and the limitations you put on it are not based on reality. :pfff: 
m
0
l
a c 229 U Graphics card
March 6, 2010 11:01:50 PM

TheGreatGrapeApe said:
Is your reading comprehension or even visual skills that Bad? Wow you really do suck at researching your own statements. :lol: 


Now, as before, you just don't read before responding. When things don't make sense, and the same article contradicts itself, I take the time to read the whole article....and look for corroborating information. Again, did you bother to note that the article contradicts itself....and after all the trouble I went through to point that out in advance for you so you wouldn't walk right into this one.

Your link with the graph

http://www.tomshardware.com/reviews/radeon-hd-5830,2564...

Same article you linked to, few pages ahead, page 4

http://www.tomshardware.com/reviews/radeon-hd-5830,2564...

DiRT 2
Version 1.0.0, Custom THG Benchmark
Run 1: Ultra High Settings, No AA, DirectX 9
Run 2: Ultra High Settings, No AA, DirectX 11

hmmmmm....the author states his test configuration page that "No AA" and lookie there.... it doesn't match it's graph what happened there.....he says two different things in two different spots in the same article. I am not saying 4AA is right, I am not saying No AA is right, what I am saying is that until it's resolved, the info is suspect.

So what do we have ?

1. THG article No. 1 says 29.6/ 33.4 w/ no AA
2. THG article No. 2 (dated 02-25-10) says it tested with no AA but yet its graph shows 4AA w/ a number (29.5)/33.5) that just missed what's in Article 1 above.
3. HH article No. 1 says 15 w/ 4AA (old drivers)
4. HH article No. 2 (dated 02-25-10) says 5770 does 30 fps w/ 2AA with latest drivers. Published same day w/ same drivers as your reference.

Obviously THG article No. 2 has conflicting information...first it says it used no AA then it says it did . Now we can take a myopic view and ignore the conflicting information within the article or we can compare the conflicting info in the article w/ the other information available to us. The evidence before us seems to suggests that the test configuration is right and the graph is wrong, however it could be either one. I am not as all fired up to rely on the stated test configuration page as you seem to be to rely on the graph. My only point is, until the conflict is addressed, it's still up in the air.

However, the fact that HH has it at 30 fps w/ 2AA kinda rules out the possibility that it can do 30.0 in 4AA. Your proof simply is not that. All we still have is two references, one of which we know has a conlicting information and is therefore in doubt at this point and another which shows that the 5770 can do 30 fps at 2AA ..... most would say that that is enough to preclude it from matching that number in 4AA.

Ya can't just pick and choose the references that support your position. If you were confident in your data, you wouldn't have to try and bolster a weak position w/ insults and name calling.

If it turns out that your graph is right, and the test configuration is wrong, whaddya got ?

1. Conflicting information on 5770 w/ latest drivers saying Dirt2 does / does not maintain 30 fps at 4AA.
2. Nothing showing that the 5770 can do better than 15 fps at max settings in STALKER
3. Nothing showing that the 5770 can do better than 30 fps at max settings in BattleForge
4. Nothing showing that the 5770 can do better than 21 fps at max settings in Alien Vs Predator
5. Nothing showing that the 5770 can do better than 30 fps at max settings in Battlefield: Bad Company 2

When ya can show me that the 5770 can do 30 fps with all settings maxed in all of those games, you will have proved that driver improvements have made the 5770 a worthwhile choice for today's DX11 games. It's obviously not there yet....

But we still haven't gotten to my point which is the "crux of the matter".

Who is building a box today with the ultimate goal of being able to play today's games ? I build boxes with the ultimate goal of doing what it's owner wants it to do for at least two years after I build it. That's when it's CPU / GPU upgrade time. I can't in good conscience tell anyone that the 5770 will do that for them.

m
0
l
a c 229 U Graphics card
March 6, 2010 11:29:51 PM

ct1615 said:
what industry? can you link it? who voted on this measure?


THG ..... Yes I can.....THG members

http://www.tomshardware.com/forum/262992-33-person-shoo...

As you can see, well over 50 % of the voting THG members said 50 fps.....taking the poll posters formula that min fps is 60% of avg fps, that 50 fps avg works out to a 30 fps minimum.

Personally, I have usually taken the position that 24-25 fps is the point that I start to get annoyed. I have often been taken to task on it here on THG over the last ....geez....will be 15 years soon. You can do your own search on "minimum fps" here or anywhere else on the web and you'll see claims from 30 to 60 and even some peeps who claim they can notice when it drops below 100. Here's some of the examples outta the top 5 hits I just got.

45 http://www.tomshardware.com/forum/211594-33-gaming-movi...
60 http://www.tomshardware.com/forum/249735-33-matter

Part of goes to the fact that most review sites use benchmark scripts to test and publish the results.....GFX card vendors work very hard on optimizing their drivers for these specific scripts. A a result, 30 fps in a scripted benchmark portion means you're likely to experience significantly lesser numbers than what you will experience at various points in the game outside of the optimized script portion.

My experience playing Crysis Warhead on my sons box at full settings went quite better than I had expected ......however, towards the end, there were two spots where it stuttered (airport tower and just before final boss) which I found quite frustrating. Each spot lasted < 30 seconds but having the hardware choice "going in" I woulda opted for more GFX oomph.


m
0
l
a c 229 U Graphics card
March 6, 2010 11:51:06 PM

TheGreatGrapeApe said:
Which still goes to the original point that it can in fact do 30fps, even by your own anal standards.


You're still ignoring the hardware heaven article which says it can't....

Quote:
You can pretend I tried to hide something, but it's all there for everyone to see and always has been.


You posted the perty picture previously, but if you posted the link as I had repeatedly requested, I missed it.

Quote:
The rest is a much later addendum


I didn't feel the need to post links ad nauseum saying the same thing over and over. To make the point that it couldn't do 30 fps, I only needed one instance to make my point......now that you pointed out that ATI has optimized the driver for Dirt2, again, all I need is another DX11 game to make my point.....all you've shown that there's conflicting evidence that the 5770 can do 30 fps in 20% of the DX11 games.

Quote:
especially for AvP which isn't even optimized yet,


Well I am sure ATI will get around to optimizing their driver for the AvP benchmark scripts.....which is why I put more reliance on hardwareheaven tests which doesn't use scripts.

But still, again, my point simply is that I am not recommending the 5770 because if it can only do 34 fps (to use your THG source containing the conflicting info as to whether AA was used) after several months of Dirt2 optimizations, what the heck will it do with the big games of XMas 2011 ? I don't build a machine in March 2010 to get someone thru what they wanna play thru April 2010....I am building to get them at least through March 2012. I don't see the 57xx series doing that and you have never addressed the central point of this whole issue.

Will a 5770 get you thru the games of XMas 2010 @ 1920 x 1200 and full settings ?
Will a 5770 get you thru the games of XMas 2011 @ 1920 x 1200 and full settings ?
Will a 5770 get you thru the games of XMas 2012 @ 1920 x 1200 and full settings ?
Will a 5770 get you thru the games of XMas 2013 @ 1920 x 1200 and full settings ?

That is the decision people are thinking when building a new machine. Personally, I am only interested in the 2nd one in bold and I don't believe that the 5770 can not do that.
m
0
l
a c 235 U Graphics card
March 6, 2010 11:59:53 PM

JackNaylorPE said:
THG ..... Yes I can.....THG members

http://www.tomshardware.com/forum/262992-33-person-shoo...



Oh I'm sorry a poll on Tom' Forum with 62 votes now speaks for the entire PC gaming "industry" * (hardware, software, and gamers). That is very interesting. I figured you could at least get the American Dental Association to agree with you since 4 out of 5 dentist agree on everything but you decided on a Poll on a gamers forum with 62 votes (7 of which are blank).

Thank you very much for helping to prove to everyone here the fact that you truly are an idiot. You single handily did a spectacular job on that. Certainly there is some small village in Europe that is desperate for man of your "skills", no need to constantly remind us with your posts.


* quoted from JackNaylor's post above "30 fps is the defacto accepted industry .."
m
0
l
March 7, 2010 12:09:37 AM

Everyone has different standard in their mind.

As long as I don't feel any lag in game with a certain GFX, I would consider it as an awesome card.
m
0
l
a b U Graphics card
March 7, 2010 1:13:34 AM

About AVP DX11 1920x1080, Maximum settings, 2xAA. Did a test run today to see if what was being said here was ture, I disabled crossfire. My single 5770 never dropped below 32 FPS, except in the scenario where the predator first appears and it is rendering many out-door things including plants, rain, lightning, and more. Where it dropped to 23.5 FPS for all of around 3 seconds. It was perfectly fine playing, and the only other place I ever saw FPS dropping through-out the Temple were when there were multiple lightsources being rendered from the flashlight. It went down to 27 FPS, which was acceptable. Max FPS 53, minimum 23.5 average 32 FPS. So AVP was completely playable with some eye candy.
m
0
l
!