Summary: To boost its performance, NVIDIA's activated 24 shaders in their revised GTX 260 GPU. In this article we take a look at the new GPU and two factory OC'ed cards based on it. See how all the cards stack up against each other in this article!
With 240 stream processors clocked at nearly 1.3GHz, a 512-bit memory interface providing over 141GB/sec of memory bandwidth and 1.4 billion transistors, NVIDIA’s GeForce GTX 280 GPU boasted impressive specs on paper. Its cheaper sibling, the GeForce GTX 260 is also a pretty formidable graphics processor featuring 192 shaders and a slightly narrower 448-bit memory interface capable of delivering 111.9GB/sec of peak memory bandwidth to the GPU. In practice with actual games, the GeForce GTX 260 ran 19-27% faster than the GeForce 8800 GTX in our testing, although there were a few cases where it was nearly 40% faster than the 8800 GTX. The GeForce GTX 280 generally ran around 15% faster than the GTX 260.
All this sounds good in theory right? It turns out ATI had a game changer with their RV770 architecture.
Going into 2008, everyone underestimated the capabilities of ATI’s RV770 GPU. We all knew ATI had two SKUs planned for launch, a 4850 and a 4870, and that the architecture would boast more shaders than RV670, but performance expectations were rather mild. It was expected that the 4850 SKU would run a little bit faster than NVIDIA’s GeForce 8800 GT, while the 4870 was gunning for the GeForce 9800 GTX. Estimates pegged it to run about 10-20% faster than the 9800 GTX.
Ultimately those estimates from January 2008 were way off: the Radeon 4850 was priced like an 8800 GT, but actually outperforms the bone stock GeForce 9800 GTX in most cases, while the Radeon 4870 is more of a competitor for the GeForce GTX 260 than anything else NVIDIA offers while boasting a $300 price tag.
As this became more apparent NVIDIA was forced to respond. The MSRP of the GeForce GTX 260 went from $450 ahead of the GTX launch to $400 on launch day. Then over the July 4th holiday NVIDIA cut GTX 260 prices again, with the GPU falling to $300-$330. Today GeForce GTX 260 cards can now be found selling online for less than $250 after rebate. Meanwhile the GeForce GTX 280 went from $650 on launch day to less than $400 after rebate today.
Obviously with a 1.4 billion transistor part sporting a 404-bit or 512-bit memory interface with 896MB or 1GB of memory and a massive die size that’s nearly 600 square millimeters, these cards aren’t cheap for NVIDIA to produce. NVIDIA is now scrambling to bring the GPU’s die size down by going from TSMC’s 65-nm manufacturing process to their smaller 55-nm process, and it’s believed that this 55-nm GT200 GPU will also use a narrower 256-bit memory interface with GDDR5 memory, similar to the arrangement ATI employs with their Radeon 4870. These two steps should reduce GT200’s manufacturing costs for NVIDIA. But in the meantime until these 55-nm GPUs are ready, NVIDIA’s strategy is to improve GT200 performance. More specifically the GT200 GPU that’s hurting their margins the worst: the GeForce GTX 260.
By improving GTX 260’s performance they should be able to charge a higher price for this enhanced GTX 260 GPU variant. As a result of the higher price, profit margins should improve, as NVIDIA’s production cost remains fixed.
That’s the theory at least. But is NVIDIA’s enhanced GTX 260 GPU capable of delivering enough performance to justify a higher price tag? That’s the question we’re here today to answer.
So what improvements has NVIDIA implemented to spice up the GeForce GTX 260’s performance? The key addition is more shaders.
If you recall the architecture of the original GeForce GTX 260, NVIDIA disabled two texture processing clusters (TPCs) when compared to GeForce GTX 280, leaving a total of 192 active stream processors. For their revised GeForce GTX 260 GPU, NVIDIA reactivates one of those TPCs, leaving a grand total of 216 active stream processors, a figure which is just 24 shaders shy of the GeForce GTX 280.
The revised GeForce GTX 260 also sports more texture filtering units, 72 versus 64 in the original GTX 260.
Everywhere else the architecture is the same as the original GeForce GTX 260. Both GPUs run at the same clock speeds and have the same 448-bit memory interface. They’re also both built on TSMC’s 65-nm manufacturing process, and according to NVIDIA the max board power between the two cards remains the same 182W.
The following chart summarizes the differences between the two GTX 260 GPUs, and how they stack up against the flagship GTX 280:
Unfortunately NVIDIA hasn’t come up with a new name for these revised 216 stream processor GeForce GTX 260 GPUs. The new chip is officially still designated as the GeForce GTX 260, as technically it’s the exact same GPU NVIDIA’s been producing for the past few months, only now they only disable one TPC instead of two.
Instead it’s up to NVIDIA’s board partners to properly distinguish their 216 shader GeForce GTX 260 cards from their original 192 shader parts. EVGA calls their boards the “GeForce GTX 260 Core 216”, while BFG goes with “GeForce GTX 260 MAXCORE”.
We really hope this doesn’t cause confusion, but with the lack of a consistent name for these revised GTX 260 GPUs we don’t see how it won’t confuse the mainstream consumer who doesn’t follow the industry everyday. After all, every GeForce board partner is going to have a different name for these 216 shader parts. This is one area where we believe NVIDIA really should consider rethinking their strategy. GTX260+ would be a more desirable designation than nothing at all.
If you’ve already got a GeForce GTX 260 and would like to purchase another for SLI, we can confirm that the new 216 shader GTX 260 boards are 100% compatible with the 192-shader GTX 260, allowing both GPUs to be combined together for SLI. Each board will run with all its shaders enabled, giving you a grand total of 408 shaders for the SLI system.
The retail cards
Physically the new 216 shader GeForce GTX 260 boards are identical to their predecessors. NVIDIA has made no adjustments to the reference board design or its cooling. Everything is the same, and we’ve been told that other than clock speed differences you likely won’t see any of NVIDIA’s board partners deviate from the reference board design.
Besides the OCX SKU, BFG will also be producing two additional 216 shader GeForce GTX 260 cards, the BFG GeForce GTX 260 OC MAXCORE and the GTX 260 OC2 MAXCORE. The OC board runs at 590MHz core/1296MHz shaders/999MHz memory, while the OC2 board is clocked a little higher, running at 630MHz core/1350MHz shaders/1063MHz memory (2126MHz effective). The OC and OC2 cards will carry an MSRP of $299.99 and $319.99 respectively.
The other GeForce GTX 260 board we received comes from EVGA. EVGA has two SKUs planned for the revised GTX 260, the EVGA GeForce GTX 260 Core 216 and the GeForce GTX 260 Core 216 Superclocked. The first card runs at the standard GTX 260 clock speeds while the Superclocked board is OC’ed to run at 626MHz core/1350MHz shaders/1053MHz memory (2106MHz effective). EVGA’s stock GeForce GTX 260 Core 216 is priced at $279 while we’ve been told that the Superclocked model will sell for $299.
Intel Core 2 Extreme QX9770
EVGA nForce 790i Ultra SLI
4GB OCZ Platinum DDR2-1333
GeForce GTX 280
GeForce GTX 260
EVGA GeForce GTX 260 Core 216
BFG GeForce GTX 260 OCX MAXCORE
ATI Radeon 4870
300GB Western Digital Caviar SE
Windows Vista Ultimate 64-bit w/Service Pack 1
Company of Heroes 1.71
Crysis High – Direct3D
The stock 216 shader GeForce GTX 260 runs anywhere from 4-9% faster than the original GTX 260. There were a couple of cases where that margin increased to 10%, and our results with Call of Duty 4 bumped that figure up to 12%, but the majority of our tests between the two GPUs were much closer than that.
Considering that the GTX 280 generally runs 15-20% faster than the stock 192-shader GeForce GTX 260, NVIDIA’s revised 216 shader GTX 260 GPU essentially knocks that margin in half.
When you then combine the extra shaders with factory OC’ing, that deficit shrinks even further. BFG’s GeForce GTX 260 OCX MAXCORE actually pulled even with, or outran the stock GTX 280 in a few cases. These kinds of performance numbers may certainly have some potential GeForce GTX 280 shoppers thinking twice about which card to purchase, particularly if you’re on a budget.
If you really want to be thrifty, EVGA’s GTX 260 Core 216 SuperClocked board combines OC’ing with a $300 price tag. You could then manually OC the board if you want a little more performance.
Of course, don’t forget that NVIDIA’s GTX 280 can also be overclocked.
ATI’s Radeon 4870 continues to put up a strong showing also. It performed neck-and-neck with the 216 shader GTX 260 in Crysis, and actually outran all GeForce boards in Devil May Cry 4. The Radeon 4870 also continued to outperform the stock 216 shader GeForce GTX 260 in most of our tests with 8xAA.
In addition, ATI’s board partners are busily cranking out their second wave of Radeon 4870 cards. These boards are shipping with enhanced coolers and/or are also OC’ed for added performance.
Ultimately we think the 216 shader GeForce GTX 260 is a step in the right direction for NVIDIA, but it’s a stopgap solution until they can get to 55-nm. Personally we think NVIDIA would’ve been better served if they’d bumped up the clocks to go along with the additional shaders, but obviously they didn’t want the card to compete too closely with the GTX 280. Of course, to counter this they could’ve bumped up the clocks for on a revised GTX 280 card as well, but that’s a discussion best left for the comments.
|© Copyright 2003 FS Media, Inc.|