ATi Radeon 5770 Performance Test
by BFG10K on Oct.23, 2009, under Articles, Technology
Hardware
- nVidia GeForce GTX285 (1 GB, reference clocks).
- nVidia GeForce GTX260+ (896 MB, reference clocks, 216 core edition).
- ATi Radeon 5770 (1 GB, reference clocks).
- ATi Radeon 4850 (512 MB, reference clocks).
- Intel Core 2 Duo E6850 (reference 3 GHz clock).
- 4 GB DDR2-800 RAM (4×1 GB, dual-channel).
- Gigabyte GA-G33M-DS2R (Intel G33 chipset, F7 BIOS).
- Creative X-Fi XtremeMusic.
- 30” HP LP3065 (maximum resolution 2560×1600).
Software
- Windows XP 32 bit SP3
- nVidia driver 191.07, high quality filtering, all optimizations off, LOD clamp enabled, hardware PhysX disabled.
- ATi Catalyst 9.9 (4850), Catalyst 9.10rc CD driver (5770), high quality mip-mapping, Catalyst AI on standard
- DirectX August 2009.
- All games patched to their latest versions.
Settings
- 16xAF forced in the driver, vsync forced off in the driver.
- AA forced either through the driver or enabled in-game, whichever works better.
- TrAA/AAA disabled.
- Highest quality sound (stereo) used in all games.
- All results show an average framerate.
12 comments for this entry:
January 5th, 2010 on 2:57 am
When i had a 5870, I would always leave the fan speed at 45% and i never touched it after that. temperature would never go above sixty degrees celscius despite being overclocked beyond 1 ghz, i dont know if the 5770 is any different. apparently, my other friend who got an xfx 5870 (i had sapphire) was getting temperatures of 70 degrees on load with 100% fan, no over clock.
In my opinion, every heatsink is diffent in a way, it might be the thermal paste.
still, its a nice review to have a whole variety of games. this makes choosing a graphics card much easier for people.
January 6th, 2010 on 1:59 am
Yes, I like testing a lot of different games (especially older titles) using a mixture of different settings. This gives a more accurate representation of how the cards behave in gaming.
January 17th, 2010 on 11:59 am
This review was very helpful and informative, thank you very much.
January 30th, 2010 on 6:48 pm
Very helpful review, thank you very much. I was looking to change my old GTS 250 (512MB) and I think the HD 5770 (probably a Sapphire Vapor-X edition) would still be a little good upgrade with DX11 support and Eyefinity support (I actually need a 8400GS that I use for PhysX to plug in a 3rd monitor). Moreover, my board doesn’t support SLI but does support Crossfire (P45 chipset).
February 7th, 2010 on 11:15 pm
Outliers are the problem here, when comparing the HD 5770 and the GTX 260. There are some massive outliers i can see straight off like Ut 2004 and Fear 2 for the HD 5770. I thought both of those games were terrible, so i wont be playing them. The would drag the average down to a more (for me) comparable level.
February 13th, 2010 on 2:01 am
Gernads and Strife, that’s exactly why I like testing a lot of games. The more games that are tested, the more readers can pick and choose based on what titles they’re interested in.
Also UT2004 doesn’t influence the ranking against nVidia because that game was only tested against a 4850 using 24xAA.
March 23rd, 2010 on 6:09 am
Your results are skewed and there is a large amount of abnormality for your Older OpenGL game results.
A GTX 275 is capable of running Quake 2 at at 1920×1200 80FPS WITH 8xSQ Hybrid AA enabled.
The wolfenstien scores are also questionably, as are doom and prey.
I expect the configuration was incorrect and you were infact benching the games in 16bpp, which will infact be much slower on new hardware, then if you were to use 32bpp.
March 23rd, 2010 on 9:07 am
Squall:
I’m not running Quake 2, I’m running a source port. There’s a big chunk of text pointing that out, right above the results.
As for the rest of the OpenGL benchmarks, there’s nothing questionable about them. ATi’s exceptional performance with the Doom 3 engine is a known fact, dating back to the X1950 XTX days. Also known is nVidia’s large performance hit with 8xAA on G80 parts (and later) in said engine.
April 5th, 2010 on 2:43 pm
one of the best reviews i have seen for this card but the results would vary. it depends on what model of card you get eg. evga 260 or msi twin frozer 260 (big difference) the overclocked vapour x version of the card over clocks alot better than the ati 5770 and mine runs at 28 idle and 46- 58 full load and i got for $15 less only games i have come along that i cant max at playable frame rates are crysis and gta iv but awsome review anyways. thanks also 2 2 vapour x are running the same as a stock gtx295 so i recomend it.
June 6th, 2010 on 6:51 pm
Please Help me i am not using more than one monitor so which is BEST for GAMEs ? which is more powerful
4890 has 256 bit but 5770 has 128 so why 5770 should be a better VGA ?
July 3rd, 2010 on 10:13 am
Sam,
Just read the comments above and you will see what card is best. They are situational. It depends on what you are using your Video Card for primarily. If you want more indepth information email me at nicholas.seigel@yahoo.com
Thanks,
Nic
MCSE
PC GAMER
High Performance PC Enthusiast
Tehnology Nutcase
July 30th, 2010 on 1:35 am
Thnx, those numbers were very helpful !!