TweakTown
NordicHardware
VR-Zone
Check out the reviews.
Xbit Labs
Hot Hardware
Hexus
FiringSquad
LegitReviews
ComputerShopper
As we said so many times before, AMD loves multi-GPU and just keeps pushing for ways of combining several GPUs to compete with NVIDIA's more powerful single-GPU solutions. Earlier this month we reported about how AMD was planning to launch a new dual-GPU card called Radeon HD 3850X2, and that it would arrive in late Q2, just in time for Computex. The exact specifications were unknown at the time, but it seems one of our forum users was right when he guessed that the memory might be cut in half.
Radeon HD 3850X2 will sport two RV670 cores running at a yet undetermined frequency, but most likely 668MHz. The memory buffer will be two sets of 256MB GDDR3 memory running at "only" 1.8GHz. However, overclocked versions and even versions with more memory should be expected from partners, which means that you might be able to find Radeon HD 3850X2 cards running at ~750-800MHz core and 1GB memory, perhaps also overclocked.
With the excellent yields, they can keep launching new cards that uses the same core, but in different ways and just start piling up money from the increased sales. The price will be slightly lower, somewhere in the $299-349 range.
http://www.nordichardware.com/news,7561.html
http://www.tweaktown.com/news/9202/nvidia_geforce_9800gtx_totally_exposed/index.html
At least on paper, it looks like nothing more than an overclocked G92 8800GT 512MB. The price tag is expected to be around $350 USD which is a reasonable price but if you were expecting a new card with ultra high performance improvements, it doesn’t seem like that will be the case at this stage and from what we have seen and heard so far.As you would expect, we have 9800GTX samples already on the way and testing will begin very soon!
If Intel gets this right (on paper at least it seems good), then the entire graphics industry will get a shake up. Nvidia and AMD have scoffed at Intel’s revelations, saying that nothing is new, and they have those capabilities already in their products. Intel on the other hand feels they have the correct formula to make some serious inroads to the GPU market with this new offering.Some of the key differences are as Intel says the “triangle/rasterisation” and “rigid pipeline architecture”. These characteristics of ATI/AMD and Nvidia GPU’s are holding back the capability of those GPU’s to perform as both CPU and GPU together, and also further complicate the physics processing options.Intel claims their product would most likely be the best choice for the dual capabilities as well as for possible physics and other computational co-processing options, more so than the current solutions provided by ATI and Nvidia.You can read all about it at the Custom PC website.
GeForce 9800GT is the successor the quite popular performance segment card GeForce 8800GT. The latter is based on the G92-270 core, and has been stripped of one shader cluster, which means that it has only 112 shader processors. The price and relative overclocking potential of the card has made it a sales success and one of the most common cards among mid-range benchmarkers. The GeForce 9800GT has been said to be quite similar and according to the review published by Chinese it168.com, it is.
The card has 112 shaders operating at 1625MHz, GPU running at 650MHz and memories clocked at 2000MHz, but the big difference between 8800GT and 9800GT is suppose to be the 3-way SLI support. The lack of 3-way SLI support is something we really missed with GeForce 8800GT, and it's nice to see that NVIDIA listened when we pointed this out.
Other than that, there are the same differences between all GeForce 8 series and 9 series cards; new and improved PCB and better power circuitry.
Written by Andreas G 21 March 2008 21:55
In the East you can find a card called Radeon HD 3690. Unlike the other cards of the Radeon HD 3600 series, 3690 is based on the RV670 core, but the memory bus has been cut in half, to 128-bit and it has only 256MB of memory. Other than that they are identical to the regular RV670-based cards. But Radeon HD 3690 is not an official card by AMD and the naming was something partners came up with. Industry sources has now told HKEPC that AMD will launch an identical card next month, but it will instead name it Radeon HD 3830.
The card is suppose to go up against the GeForce 9600GT, which we find a bit amusing as Radeon HD 3850 is several tens of dollars cheaper and 3870 costs pretty much the same, and they both perform quite similar to 9600GT.
However, overclocked versions with more memory might be able to do better, something sources have been saying should arrive shortly after the launch. MSRP will be in the $120 range, the same as GeForce 9600GT.
Without a doubt, this is a way for AMD to dump more RV670 chips onto the market. We can imagine that the fabs have been pumping out chips at high fields for some time now and that AMD is now trying to figure out the best ways to hurt NVIDIA and get rid of excess stock at the same time.
Tempest is the latest creation from case manufacturer NZXT. NZXT is known for its gaming-oriented case and focus on building cases that can house the most powerful hardware. The Tempest is no exception as NZXT proclaims it to be The Airflow King. Two 120mm fans in the front, two 140mm in the roof, one on the side and one in the back is the reason for this. This is further strengthened by the predrilled holes for better cable management to eliminate any unnecessary obstruction of the airflow.
Nehalem is Intel's next microarchitecture and it should arrive sometime before the end of the year. Intel has displayed Nehalem-based systems at IDF and even though no specific performance indications have been released, seeing a dual-core workstation with two native quad-cores running sixteen simultaneous threads is simply impressive. Even though we still don't know much more since Intel decided to unveil more details on Nehalem a few days ago, we now have pictures of a pre-production motherboard codenamed Smackdown.
Perhaps not the most exciting pictures of a motherboard, but it's nice to see that things are moving forward at Intel.
There has been a lot of buzz around the web regarding an article published by W1zzard over at TechPowerUp. He found an irregularity when testing the just launched GeForce 9600GT. It seemed that the card would perform exceptionally better when it was running with an overclocked PCIe bus, which isn't normal under these circumstances. He investigated it further and found that it seemed like the card used the PCIe frequency as a reference crystal, instead of the on-board physical crystal. A follow-up investigating the oddity was also posted.
The problem with this isn't so much that the card overclocks with the PCIe bus, it's actually quite nifty, but that the increased frequency wasn't reported by the drivers. The card seemed to operate at default frequency when it was not. People have been wondering why NVIDIA didn't reveal this to people reviewing the card, as they may have been lured into making the card look better than it was. That would be the paranoid angle of it, but right now it's the one dominating the discussions.
We still don't know why this information was omitted. It might just have been some sort of miscommunication at NVIDIA, because it has now made an official response saying that the card does indeed have two crystals. One on-board 27MHz crystal and one crystal which is connected to the PCIe bus. The things is, GeForce 9600GT isn't the only card that behaves this way. TechPowerUp discovered that GeForce 9800GX2 behaves the same way, and chances are that the rest of the GeForce 9 series cards do too.