Forum Home
Press F1
 
Thread ID: 91765 2008-07-18 03:04:00 GFX Card Recommendations a helpless random (13059) Press F1
Post ID Timestamp Content User
690016 2008-07-18 10:41:00 That's probably the best card you're going to get for the money.

I'd be happy with it.
Yes but as The Bananamonkey said that is probably the best card i can get for my price range and i can afford the 4850 and it looks good so ill go for it. 		a helpless random (13059)
690017 2008-07-18 22:30:00 Don't worry about stream processing units or anything like that, but it's fair to assume that higher is better.

I always reference any hardware choices through TomsHardware (www.tomshardware.com). They have a section called 'Best Graphics Cards for the Money' which I usually trust.

Note that the absolute best they could offer was two of these suckers. 		Thebananamonkey (7741)
690018 2008-07-19 03:13:00 Hi, i have found a 1GB card for a similar price here: pbtech.co.nz
is that better? 		a helpless random (13059)
690019 2008-07-19 03:34:00 What type of monitor will you be using? If it's an LCD under 24" I don't think a 1GB will really be of much use, as you wont be playing at a high enough resolution to need it..

I'd say go with the 4850, it seems to be be as good, if not better than a Nvidia 9800 GTX from what I've read. A 1GB 9600GT will not be able to compete. 		Sherman (9181)
690020 2008-07-19 06:50:00 Thanks, ill buy the 4850, i just couldnt not think about a 1GB GFX card:cool: but im using a 17 inch. Thanks a helpless random (13059)
690021 2008-07-19 11:20:00 What a crappy website, the 'Details' section really didn't tell me anything about the card.

"# 956 million transistors on 55nm fabrication process"

So!? Does it do what it's supposed to do? Yes? Ok then...

"# 800 stream processing units"

What the hell is a stream processing unit? :confused:

8800GTX number of transistors: 681 million
As a comparison.....

Without getting deeply in technical matters which seem to bother you - the more transistors - the better the performance.
And
Stream Processing:

computer programming paradigm, related to SIMD, that allows some applications to more easily exploit a limited form of parallel processing. Such applications can use multiple computational units, such as the floating point units on a GPU, without explicitly managing allocation, synchronization, or communication among those units.

The stream processing paradigm simplifies parallel software and hardware by restricting the parallel computation that can be performed. Given a set of data (a stream), a series of operations (kernel functions) are applied to each element in the stream. Uniform streaming, where one kernel function is applied to all elements in the stream, is typical. Kernel functions are usually pipelined, and local on-chip memory is reused to minimize external memory bandwidth. Since the kernel and stream abstractions expose data dependencies, compiler tools can fully automate and optimize on-chip management tasks. Stream processing hardware can use scoreboarding, for example, to launch DMAs at runtime, when dependencies become known.Stream processing is especially suitable for applications that exhibit three application characteristics[citation needed]:

* Compute Intensity, the number of arithmetic operations per I/O or global memory reference. In many signal processing applications today it is well over 50:1 and increasing with algorithmic complexity.
* Data Parallelism exists in a kernel if the same function is applied to all records of an input stream and a number of records can be processed simultaneously without waiting for results from previous records.
* Data Locality is a specific type of temporal locality common in signal and media processing applications where data is produced once, read once or twice later in the application, and never read again. Intermediate streams passed between kernels as well as intermediate data within kernel functions can capture this locality directly using the stream processing programming model. 		pctek (84)
1 2