System Set-Up and Software Used

This exclusive look today we get the chance of looking at the new mid range level card the FireGL V5600.   When we were first advised of this new cards arrival here and what it had hidden up its sleeve we decided to run it against its predecessor in the mid range the FireGL V7200 with a 256 bit controller and 41.6GB/s memory bandwidth and see just how much more performance would be gained.   We have already seen what the new FireGL V3600 could do!   A 128 Bit controller v 256 Bit and 35GB/s Bandwidth v 41.6GB/s this will be interesting set of tests as the new FireGL V5600 has on specification sheets 15 percent less Bandwidth than its more powerful elder!

We put together a system which would be found in most mid range level workstation scenarios, this gives us the chance to properly test Supermicro’s new socket 775 mainboards which supports the new P33 Intel Chipset and of course the new Intel ICH 9 chipset . 

The mainboards, bios settings were left at default and it automatically adjusted the clock speed settings and voltages to facilitate the CPU’s and Memory to maintain the higher functionality which the mainboard has to offer.

Hardware Suppliers for the Review 

Supermicro for the supply of the new Mainboard shown, Corsair, for the supply of their Dominator memory, Western Digital for the SATA Hard Disc, AMD for the supply of the FireGL V5600 professional graphics card, Man and Machine for the supply of the and FireGL V7200, and finally but by no means least; Akasa for a majority peripherals required for this review.

With the support and help of all the companies involved this has been a cracking first look a AMD’s new FireGL card.   It gives you the reader a choice on upgrade paths that many of you look for here with the performance ratio’s that a system has to provide.
 
Systems Integrators, OEM’s and VAR’s should contact all companies directly for pricing and availability of all components. Members of the public should contact their respective suppliers requesting the parts directly.

Diskeeper can be purchased from most leading Distributors and software retail outlets. Or those of you wishing a try before buy, a 30 day trial version maybe downloaded direct from Diskeeper Europe. We highly recommend the use of this software as performance increase can be quite considerable in many places. 

Benchmarks and Software Used (32 Bit Mode) 

The plethora of benchmarks available in this arena; one could spend all day in producing results if the time permitted, paralysis by analysis!  Sadly the clock ticked hard and fast against us therefore we did only have a very short time in which to complete this basic but factual review.

Microsoft Windows XP Professional SP2
Diskeeper 2007 Professional Premium Edition
SiSoftware XI Professional SP4
HD Tach Journalists Edition 3.0.4.0
Cinebench 9.5 & Cinebench 10 *NEW*
** SPECviewperf® 10.0

** So that readers will be able to see how much performance increase has been gained throughout the range SPECviewperf® tests by the individual cards.   We have split up into their individual test categories from each test system and run.   Therefore one page has been dedicated for the Multi Threading Test and the Overall SPEC Viewperf and Full Scene Antialiasing benchmark results from each of the mainboards/system supplied.   This will give the reader an idea of what to expect when they plan in purchase. 

To complete all of the tests on the systems has taken some considerable time to complete. As a point of note it can take at least a full 3 – 4 hours to properly complete a complete “run all tests” depending on the system under test.   However, it does feed back to the end users considerable amounts of information for digestion especially within the Composites and Multisampling Results.  

Each card completed the tests etc on a clean system hard drive to ensure that no residue drivers were left installed.   A test/render has been completed many times over different periods of the system uptime.   Maintaining the fair play rules of SPECviewperf® we did not manipulate any of this mainboards functionality and left it at its default settings in which it was supplied.   All tests have been conducted at 1280 X 1024 @ 60Hz in 32 bit colour. Results that have been shown on this review are from the application/benchmarks first run in accordance with the SPECviewperf® fair play rules.

The results speak very clearly for themselves as a picture paints a thousand words; so there is no need for a running commentary of the results gained from each system.  “Run Logs” the SPECviewperf® tests are available to those of you who wish to see them.    Please e-mail the administrator here who will forward them onto you on a first come first served basis.    This may take some time to get back to you as on first out launches of this nature we are undated with requests.