5770 = 4770?

[LeJosh]

yes, if thats a minimum and not an average lol

If the 5770's were say, £80 then it would make the 4770's obsolete. as it stands, you're paying £70% extra for a 40% increase in performance (roughly of course) which imo doesnt make it clear cut. But two of them at £80.....that'd be great if i had a decent crossfire motherboard (and not an ip35pro )

Going by that sentiment you expect atleast 1% increase per £1 spent. That's impossible in this day in age surely?

Edit - oh ffs I think I read it wrong... well I'll leave it be for funzies.

 

[Lightnix]

So, what's you're definition of 'massively faster' then? 45% is a big percentage, but is 34 fps a lot faster than 24fps? because that's a 41% increase..... i say no.

Yes, yes it is. Moreover, you could probably get away with adding some AA and run at a higher resolution with the '34'FPS card and only have it drop to 30FPS. Then the card that can only do 24FPS is still slow enough to be irritating and renders the game in comparative ugly-vision.

 

[james.miller]

A 40% increase in performance isn't massively faster?
LOL wut?

fine, be happy with a card than plays crysis warhead at at 20fps rather than 15fps. I mean come on, its 35% fasterrrr!!!!!111one11 wow huge increase. at least that's only a minimum framerate i guess!


I would accept that in the worst possible most crass inaccurate way than displaying a spinning cube at two frames a second demands twice the powah that spinning it at 1 frame per second would, but if it's still slow either way who cares? what baffles me is how excited people get over these fantastico magnificent increases in speed.


...I'm just not that easily miss-led!

Yes, yes it is. Moreover, you could probably get away with adding some AA and run at a higher resolution with the '34'FPS card and only have it drop to 30FPS. Then the card that can only do 24FPS is still slow enough to be irritating and renders the game in comparative ugly-vision.

That example was actually already at 1920x1200 4xAA, but that doesnt make an awful lot of difference. Being a minimum framerate, one is O-K (34fps) then other pushing it and probably too slow for the majority of people (24fps). Either way, one isnt massively faster than the other, no. I don't agree

 

[Emlyn_Dewar]

You know what, **** Crysis!
The 5770 is a good card, it's a better card than the 4770. It's not "basically the same chip." (This has been said in this thread I think)

I understand where you're coming from with the playability stakes but it just IS a better card.

 

[james.miller]

I agree and the chart up there ^^^^ proves that I just dont think it's faster enough to currently consider. In my opinion

 

[Phatzy]

its worth it, them extra features. at around same performance of a 4870 and in some cases beats a 4890. i think its a bargin.

 

[drunkenmaster]

I agree and the chart up there ^^^^ proves that I just dont think it's faster enough to currently consider. In my opinion

If you want to play Crysis Warhead at max settings you spend £300, but if you spend £115 on the 5770, you can play it at near max settings with it perfectly smooth, and most other games around now on basically max settings.

You not wanting to buy the card for a higher resolution and a higher setting, really doesn't mean squat, you want £300 of performance and don't want to buy a £115 card, that doesn't make that £115 slow, slow for its price or bad performance.

Using the odd game with an odd setting to get low framerate at max settings proves nothing.

There WILL be a game in the future that a 5870 will play at only 16fps, while a 5770 gives you 8fps all at max settings. Does that make the 5870 not faster than it because the 100% extra is at the low end of the scale, no, it just makes your assertion that the card isn't faster, stupid.

If you want to pretend that 40% is insignificant because ONE GAME under SPECIFIC settings is only a jump from 20 to 30fps as nothing, thats fine, of course that very same game, on lower settings, might bump that game up to 45fps on the 4770, and 65fps on the 5770, would that 40% still be insignificant then? According to you, no, infact if you went and used the lowest res, you might be looking at 100fps on the slower card, and 150fps on the faster card, still insignificant?

 

[Subtlesnake]

fine, be happy with a card than plays crysis warhead at at 20fps rather than 15fps. I mean come on, its 35% fasterrrr!!!!!111one11 wow huge increase. at least that's only a minimum framerate i guess!


I would accept that in the worst possible most crass inaccurate way than displaying a spinning cube at two frames a second demands twice the powah that spinning it at 1 frame per second would, but if it's still slow either way who cares? what baffles me is how excited people get over these fantastico magnificent increases in speed.


...I'm just not that easily miss-led!

So your point is that eventually performance differences become meaningless when you can't render a game at above 20 FPS, anyway? Well, so what? I mean that basic point applies no matter what the differential is. I mean card A may be 15 times more powerful than card B, but if you're running at some ridiculous resolution, that might equate to 15 FPS vs. 1 FPS.

Does that mean that the difference in performance doesn't matter? No, because as you dial back the complexity, you can get the playable framerates you require, and you still get 15x the detail (possible at 30 FPS).

So the mere fact that on certain benchmarks neither the 4770 or 5770 deliver playable framerates means nothing in and of itself. The same applies to the 5870, or the GTX 295. What matters is the visual quality improvement you get with the extra performance.

And getting say, a 33% higher minimum framerate, means you can raise the resolution and maybe even add AA, which is significant: giving you 30 FPS, as opposed to say, 22.5 FPS at those settings. More performance allows you to go further.

I mean what differentiates the 4850 and the 4890? 40%? Yet think about the vastly different attitudes we have towards each card, and the vastly different prices. So I think in consumer graphics, these sorts of percentage differences are significant, even if they don't mean the difference between playable and unplayable at 1920x1200 w/8xSSAA.

 

[Randell]

yes, if thats a minimum and not an average lol

rubbish an average of 48fps is perfectly acceptable in virtually every game and a long way from a 'slideshow' as long as the minimums around 25-30fps. Anyway thats beside the point. Of course the card isn't suitable for an enthusiast.

 

[Dreadhead]

The 5770 is not aimed at the gamer IMO.

I think the target audience for a single card would be the HTPC crowd.
This card sports a protected audio pathway for bitstreaming audio.
If you were to buy the only soundcard capable of this feat it would set you back between £120 and £150.

So for £110 you get this coveted feature, plus the ability to play all current games on a 1080p monitor (albeit with some options scaled back), all for 18 watts idle.

If you are a gamer first and foremost there is the 4870 for equal performance at a cheaper price and the 5850 or the 5870 at a premium.

 

[gurusan]

4770 most certainly was aimed at the gamer....midrange gamer though