backtop


Print 47 comment(s) - last by Darkskypoet.. on Feb 26 at 6:46 PM


RV740  (Source: Guru of 3D)

  (Source: Guru of 3D)
Benchmarks place it comfortably between the HD4830 and HD4850

Despite an uncharacteristically tight-lipped stance from ATI on its first 40nm parts, hardware review website Guru of 3D has managed to secure a sample of an RV740-based graphics card and put it through the paces. At first glance, the specifications suggest that it is little more than a shrunken RV770LE core, but a closer look reveals several improvements

According to the preview published today, the GPU boasts 640 shaders, 32 TMUs, and 16 ROPs; all the same as the RV770LE. However, the core and memory clocks both receive a significant bump, up from 575MHz to 650MHz and 1800MHz GDDR3 to 3200MHz GDDR5 respectively. The substantial boost in memory frequency works to offset the performance difference incurred by the smaller 128-bit memory bus on the RV740, which brings the reported math processing rate for the GPU up to 900 GigaFLOPS.

Every single benchmark result published in the review places the RV740-based card right between the existing Radeon HD4830 and HD4850 graphics cards, also from ATI. For example, at 8xMSAA and 16xAF in the popular first person shooter Left 4 Dead, the RV740 turns in 28 FPS, flanked by scores of 25 and 30FPS from the two cards at 2560x1600 resolution. The card does exhibit a significant performance drop at higher resolutions, likely the result of the 128-bit memory bus width and the relatively low 512MB of memory. Other reported benchmarks include Far Cry 2, Crysis WARHEAD, Call of Duty 5: World at War, Brothers in Arms: Hell’s Highway, and 3DMark Vantage.

MSRP for RV740-based graphics cards is expected to be $99 USD, which should make it a very attractive offering in the upper-mainstream segment. While the official name of the video card has not been released, the author of the article suspects it would be called the Radeon HD4750, which would be a logical name for a card with the given characteristics.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

new AMD Card
By frozentundra123456 on 2/25/2009 7:40:28 PM , Rating: 2
I wonder how much power this card uses. I dont really see a need for a card to fit between the 4830 and 4850 unless it saves a lot of power. I would have expected the first card at 40 nm to be a high performance one.

I really liked the simplicity of AMD's line up, basically 3 levels of cards with a lower and higher performance part in each level. Now their lineup is getting really complicated and confusing (or maybe its just me).

What did they do, hire nVidia's team in charge of naming cards?? At least they haven't yet stooped to bringing out previous generation hardware and giving model numbers to make it seem like new generation hardware (have they??).




RE: new AMD Card
By JSK on 2/25/2009 8:08:02 PM , Rating: 5
At least this is actually a "new" card and a "new" process unlike the G92 renaming debacle.

Nvidia changes stickers and AMD shrinks the die are you are trying to draw a funny comparison? This isn't even on the same level.

In all likelihood the 4830 will be phased out and maybe even the 4850 as well.


RE: new AMD Card
By PrinceGaz on 2/25/2009 9:20:10 PM , Rating: 4
quote:
I would have expected the first card at 40 nm to be a high performance one.


nVidia learned the very painful lesson from that approach back in the GeForce FX 5800 days, and ATI/AMD have always steered clear of repeating it.

Rather than use a smaller fab-process with the latest and greatest chip, it is best to use it on a low-mid range product until it is proven to have satisfactory yields on those chips, before introducing it at the high-end. Just because you have a new smaller fab process available does not necessarily make it better or faster than the chips which come off the older fabs at first. It takes time for the fab plant to become better at making parts in that process size, so for several months it is common for it to be more economical to produce equally fast or faster parts on the older fab facility.

It's best to start with the new relatively untested facility with low-end parts where there should still be lots of usable chips even with a lot of defects per die, before progressing to manufacturing the large high-end chips which require a much lower defect-rate per die for satisfactory yields.


RE: new AMD Card
By MikeMurphy on 2/26/2009 1:34:44 PM , Rating: 3
Having been an owner of a 4830 and 4850 I can say that my expectations are that the 4830 will still be a better buy for the performance-oriented purchaser than this new chip if prices reflect positioning of stock performance specs.

The RV770LE chips (4830) were binned for one of two reasons. They had defective SIMDs ...OR... they didn't clock well enough to be used as an RV770. From my research and experience the vast amount of these binned chips have defective SIMDs, which means that they clock on par with the RV770. You will find that over half of the RV770LEs clock from stock 575mhz to well over 735mhz which is a phenomenal free overclock. These chips thrive with voltage and can usually break 800mhz with a small bump in power.

Anyways, its one hell of a value chip and I suspect this new chip won't offer the same value when overclocking.


RE: new AMD Card
By coldpower27 on 2/25/2009 10:33:31 PM , Rating: 3
quote:
I would have expected the first card at 40 nm to be a high performance one.


Both ATi and nVidia got burned big time using those approaches, ATi on it's HD 2900 XT, and Nvidia on the Geforce FX 5800.

The HD 4750 makes sense, way back when Nvidia had a Geforce 6800 GS for the mid range, that was eventually replaced with the Geforce 7600 GT.

You want to ditch the wider memory interfaces to cut cost if at all possible. Hence why you get ATi doing the 256Bit only thing with the HD 4800 Series.

nVidia doesn't really need to make a new core for the mid range. The G92B is alright for now,

I wonder if a 1600 Shader Card with 64TMU is possible for 40nm... if they can double the core resource from 55nm to 40nm on the mid range card, why not the high end.


RE: new AMD Card
By Natfly on 2/26/2009 12:49:36 AM , Rating: 2
There are rumors that TSMC's 40nm process is quite borked and offers little to no power/performance gains compared to the 55nm process. I guess we'll see when the reviews eventually come out. If it's true then the only benefit of 40nm will be restricted to cost only.


RE: new AMD Card
By CyberHawk on 2/26/2009 6:55:47 AM , Rating: 4
It's just you ;)


RE: new AMD Card
By nafhan on 2/26/2009 8:07:56 AM , Rating: 2
The 4830 is partially defective 4850 with 160 stream processors turned off and a 256 bit memory interface on a 55nm process. This new card will be designed with 640 stream processors and have a 128 bit memory interface all on a 40nm process. If nothing else, this translates into huge cost savings for AMD and (hopefully) us.


“We do believe we have a moral responsibility to keep porn off the iPhone.” -- Steve Jobs











botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki