OSNews

yes you can
no matter what they will call their gpu it will still bei a radeon hd 7850 oc
just compare cores, bus, throughput and performance
but they will manufacture it in a newer process to get powerconsumption down

By this logic all ATI graphics cards in last 15 years are Rage 128 Pro but with lower manufacturing processes.

I/O these days is the bottle neck.

My ancient nforce P5N-SLI machine with a Core 2 duo 8400 machine is playing Crysis 3 just off ultra settings.

I have maxed out this machine, and the only speed improvement I got going is overclocking the CPU (which means better RAM, since the multiplier is locked on intel chips).

My motherboard was purchased 6 or 7 years ago, the CPU is a E8400 and I am running 8GB of DDR2. Both of which is ancient by today's standards.

My best upgrade to my battle-rig ... two raided SSD drives. VS2012 opens instantly from cold boot. Win8 takes a few seconds to load once I am past the BIOS screens.

Processing power, ram and latency isn't a problem. Disk I/O is.

Edited 2013-02-21 20:08 UTC

But SSD raids do not pass trim command and performance degrades over time, or did they fixed that?

Processing power of the GPU is always a problem if you want to implement nice visual effects. More graphic calls/second, the better.

Ram and latency are most definitely an issue. Just as Disk I/O is orders of magnitude slower than RAM access, RAM access is orders of magnitude slower than Cache access.

No matter how fast your storage (disk) subsystem can spit out data, it's not going to be anywhere close to as fast as getting it from memory. And ditto, getting data or instructions from cache is far superior to making the processor use it's memory controller to look up an address from some page table and do the virtual to physical lookup.

If you have enough RAM to fit all the game data into sytem RAM, so much the better. Unfortunately, SRAM is way too expensive, and even now, 128K Harvard architecture split data/instruction caches is considered pretty good. Getting access to L1 cache can take just a few clock cycles. Getting memory from system RAM takes a few hundred clock cycles generally speaking. But whereas system RAM access is measured in micro or even nano seconds, disk IO is stilled measured in milliseconds. In other words, disk access is thousands of times slower than RAM access. If you're page faulting and have to swap to disk, you're in BIG trouble in a game. The idea is you pull data from storage into RAM. RAM times are so important, that many developers don't even use malloc or new, and they pre-allocate their own heaps (memory allocation is expensive and can put the processor to sleep, not to mention the problem with memory fragmentation).

In regards to processing power, the name of the game is concurrency. Many systems are now merging the concept of CPU and GPUs so that the massively parallel processing of GPU's can work on all the highly vectorized game data (usually graphics, but physics too). But many other components will still rely on the main CPU (artificial intelligence for example). Splitting data across multiple threads of execution is very challenging.

Is a game more GPU or CPU bound? Depends on the game I suppose (I'm not a game developer, I'm a storage controller device driver developer). But again, processing and especially optimizing parallelization and concurrency will provide FAR greater beneft than disk IO.

Given AMDs track record the past many years, it's a pretty safe bet though.

Jaguar is the successor to the Bobcat core found in Brazos APUs. This is AMD's competitor to Atom, it won't be a particularly high performance part. It's supposed to be a substantial improvement over Bobcat, but it's not meant to compete with more power hungry desktop cores.


Sony has stated that the GPU is capable of 1.84 TFLOPs which is a little more than a Radeon 7850 (1.76 TFLOPS). AMD isn't introducing a radically new micro-architecture with Sea Islands so I think it's fair to say one of the high end Soutern Islands GPUs (like the 7870 GHz Edition or higher) will easily outperform it.


This is true, but it's also shared between the GPU and CPU whereas in a typical gaming PC the video card will have it's own dedicated pool of GDDR5. There are certainly advantages to a unified memory architecture. In particular, there are certain tasks that would be well suited to running on a GPU if it were not for the cost of shuffling data between main and video memory, but there are other tasks in which the greater aggregate bandwidth of a system with separate main and graphics memory wins out.

I'm not sure what the big deal is though. Apart from certain classes of problems for which the Cell was particularly well suited, the PS3 was not terribly impressive compared to PCs of the time.

a hd 7850 at 900mhz (instead of 860) has 1,84 TFLOPS