PowerColor HD5770 Video Card Review @ [H]

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,772
PowerColor HD5770 Video Card Review - Think your old GeForce GTX 260 or Radeon HD 4870 is getting a little long in the tooth? They’ve both had a great run, but now it might be time to upgrade. We see how these two video cards compare to the brand new PowerColor HD5770.

If you currently own an ATI Radeon HD 4870 or a NVIDIA GeForce GTX 260, don’t look for a huge gameplay improvement in today’s games, it just isn’t there right now. But if you are currently in the market for something new, supporting forward looking DX11 games, and the ability to run triple-display gaming at an affordable price, then the PowerColor HD5770 is where you want to be looking. The PowerColor HD5770 provides an excellent gameplay experience for the price of $165 with the future in mind.
 
they cooled only half of the memory chips!?

Yeah I noticed that as well on the pictures that is a odd design. I see plainly on the back of the card that there is uncooled Video RAM chips which is a really poor design to me. That just seems like the card will burn out that much faster.
 
Yeah I noticed that as well on the pictures that is a odd design. I see plainly on the back of the card that there is uncooled Video RAM chips which is a really poor design to me. That just seems like the card will burn out that much faster.

I do not think it will burn out faster, this is standard/stock design by AMD on HD 5770.
 
Yeah I noticed that as well on the pictures that is a odd design. I see plainly on the back of the card that there is uncooled Video RAM chips which is a really poor design to me. That just seems like the card will burn out that much faster.
Here:
http://www.hardforum.com/showthread.php?t=1469834&page=3&highlight=5750+5770+cooler

is what I was looking for.


The Memory chips on one side are cooled (because of ambient heat from the GPU?), but the other side isn't.

wait... I think I just figured out why.
 
Call me stupid .... but the GTX 260 was outperforming that thing. Why would anyone without money to burn upgrade to it at this point?
 
interesting to say the least. Brent, when you overclocked the card did you happen to try and OC the memory, esp for batman given the performance difference? ATI is adamant about the memory bandwidth not being an issue so it would be interesting to test it.

still a very nice midrange card
 
I'm sure I'm in the minority here, but I would love [H] to start testing the noise levels of the fans on these cards. I have been trying to decide whether or not I should pick one of these cards up, but I am still unclear on just how loud these cards are. Newegg reviews are ok, but I think a lot of people consider cards quiet that I would say are annoyingly loud. Any chance of [H] getting a sound test up and rolling?
 
interesting to say the least. Brent, when you overclocked the card did you happen to try and OC the memory, esp for batman given the performance difference? ATI is adamant about the memory bandwidth not being an issue so it would be interesting to test it.

still a very nice midrange card

Yes we overclocked memory, it shows that on the table - http://hardocp.com/article/2009/11/23/powercolor_hd5770_video_card_review/9

We also have an apples to apples graph on that page in Batman with and without the overclock.
 
I'm sure I'm in the minority here, but I would love [H] to start testing the noise levels of the fans on these cards. I have been trying to decide whether or not I should pick one of these cards up, but I am still unclear on just how loud these cards are. Newegg reviews are ok, but I think a lot of people consider cards quiet that I would say are annoyingly loud. Any chance of [H] getting a sound test up and rolling?

We use to include decibel readings, but that was always a crap shoot, quite frankly we aren't setup to do formal sound testing. Since we aren't set up for it properly I'm not going to half arse it. For now, subjective feedback on the noise will continue.
 
It'll be even more important on 5750, which is cheaper, I would say 5750 would be better suited for HTPC crowd. We'll be sure to mention it on those evaluations.
 
Upgrading from a 4870 or a GTX 260 to a 5770 is the most absurd notion. All of the initial reviews of the 5770 showed as much. I don't know why the article was couched in those terms.
 
We use to include decibel readings, but that was always a crap shoot, quite frankly we aren't setup to do formal sound testing. Since we aren't set up for it properly I'm not going to half arse it. For now, subjective feedback on the noise will continue.

Thanks for the reply. Is setting up a sound testing area a pretty difficult process then?
 
Thanks for the reply. Is setting up a sound testing area a pretty difficult process then?

Wouldn't be terribly difficult provided you have the extra space. Need a smallish room, some acoustic dampening materials/treatment for the walls, and test equipment. Not sure how much value it would add to subjective testing in this area though. There are different types of noise, some of which are more annoying depending on the person. For example, white(ish) noise from airflow may not be as annoying as whine from the fan itself, even though it may be louder (db-wise.) You could get some raw numbers, but there would still be a subjective element to it.
 
It seems to me that Eyefinity could hardly be called a differentiating feature when the card barely has the power to hang with an old 260GTX at standard resolutions.
 
I have to agree I just don't think the reference would convince anyone in their right mind. If you have gtx 260 or greater this is not the card to upgrade to. I know you have to review Hardware but you don't have to hawk it to us when its unnecessary do you?
 
Wouldn't be terribly difficult provided you have the extra space. Need a smallish room, some acoustic dampening materials/treatment for the walls, and test equipment. Not sure how much value it would add to subjective testing in this area though. There are different types of noise, some of which are more annoying depending on the person. For example, white(ish) noise from airflow may not be as annoying as whine from the fan itself, even though it may be louder (db-wise.) You could get some raw numbers, but there would still be a subjective element to it.

Good point. I mean you could put a recording of it up online. But then what if someone has his speakers way too loud/soft? I guess I'll just wait and hope [H] finds some way of testing sound that actually means something, much like they did with their real world benchmarking.
 
I'm curious if the GTX 260 and 4870 would be capable of higher levels of AA then the 5770 at a resolution of 1680x1050. I'm wondering if the memory bandwidth of the 5770 would handicap it at 24x AA.
 
[H], any chance we could get a review on the V2 5770 with egg cooler that was linked in the review? We've had some debate in the forums as to whether or not the cooling and noise are improved with this design. Is it a better alternative to the V1 with the batmobile shroud, or is it simply a cost cutting measure?
 
My last card was a 3870, I moved to a 5770 version 1, I have to ask - can it get more silent? I run a stock Phenom II x2 550 cooler and four system fans, live in a grim and frostbitten country (Sweden).
 
I tried a 5750 recently and I was really impressed by the power consumption and heat output but stability just wasn't there. I tried the Catalyst 9.11 drivers but there were still issues. Has anyone done any long term stability testing on these new cards with the latest drivers? I'd like to see if they can handle something simple like a handful of hours just looping 3D Mark or similar. I might have just had a bad card but there seems to be lots of complaints about the stability of the drivers presently available. Most review sites are giving these new cards the thumbs up with no mention of long term stability.
 
Yes we overclocked memory, it shows that on the table - http://hardocp.com/article/2009/11/23/powercolor_hd5770_video_card_review/9

We also have an apples to apples graph on that page in Batman with and without the overclock.

let me rephrase the question correctly. Did you happen to try overclocking just the memory to see if the performance difference was directly attributable to the memory bandwidth? (that was a nice OC BTW, wonder what you could do with a volt mod)
 
In the interest of accuracy, I'd like to point out that bales of hay may or may not make good cover, depending on their density. At best, I'd think most bales of hay would be concealment, especially if the OPFOR has anything bigger than a rifle.

On the other hand, I'm not sure how ArmA handles bales of hay. My guess would be that they'd be treated as concealment.
 
Thanx for this review. I have a 5770 arriving tomorrow, and NFSS & Arma II are both titles I was wanting to be able to play @1920x... now I know I can. :)
 
If you don't currently have a card at that performance point the HD 5770 is compelling. If you have an HD 4870 or GTX 260 (216) the HD 5770 is no upgrade. Its wishful thinking that it will ever be an upgrade at any point in time.
 
I wholeheartedly concur with the "Future Potential" portion of the conclusion. As such, I'm personally waiting until those DX11 games comes out to see whether or not DX11 will succeed where DX10 pretty much flopped. If DX11 is a big thing, that definitely will make it easier for me to pick up the HD 5770. Besides, by then drivers would be more mature and prices lowered a bit.
 
Thanks for the review!
I am curious to know, why test a mid to low end video card on such a beefy system? My person opinion here but if I owned that system I wouldn’t have purchased a 5770 for it. Anyway to get lower end systems for lower end cards thrown in the mix as well?
 
Thanks for the review!
I am curious to know, why test a mid to low end video card on such a beefy system? My person opinion here but if I owned that system I wouldn’t have purchased a 5770 for it. Anyway to get lower end systems for lower end cards thrown in the mix as well?

That's covered in the methology mate... testing the Graphics cards only, really you just read how much better/worse one is vs the other, relative not absolute.

basically take the (rest of) system out of the equation, and compare whats best in the line up - quite like this tbh.
 
That's covered in the methology mate... testing the Graphics cards only, really you just read how much better/worse one is vs the other, relative not absolute.

basically take the (rest of) system out of the equation, and compare whats best in the line up - quite like this tbh.

I understand this but it would be nice to have a low end, mid end and a high end system in the mix. [H] has already covered a reference 5770 why not expand on what I the real world user would see on general systems low, mid and high end systems. Just curious as to why this wasn’t done?
 
Last edited:
If you'll notice, they used an OC'd i7 in the last review and an OC'd QX9650 in this one. Not quite High/Mid but def Very High/High. They did this also in the 5870 review and it made the gap between the 5870 and 295 more noticable with the i7 setup, IIRC.
 
Upgrading from a 4870 or a GTX 260 to a 5770 is the most absurd notion. All of the initial reviews of the 5770 showed as much. I don't know why the article was couched in those terms.


I disagree.

If you sell your 4870 or 260 for around $100, add $50 or $75, you are in great shape.

I just did that.

I sold my 260-216 for around $128 shipped on 10-21-2009.
I picked up a XFX 5770 for around $165 shipped at Newegg.

I now have a quieter, cooler running card, which has DVI, HDMI and Display Port connectors, supports multiple displays / eyefinity, and DX11, and is just about as fast as my 260-216 on very immature drivers and older games.

In 6 months or a year I should be even happier.


How can you go wrong with that?
 
Back
Top