Single core AMD --> C2D

[sswilson]

I'm picking up an E6420 + one of those S775 Frankenstein Asrock boards (DDR1,DDR2,AGP,PCIe) tomorrow and will be doing a few comparison benchmarks between it and my current rig.

I'm going to be using my DDR1 on the Asrock & the PCIe bus on the asrock is limited to 4X so I'm mostly interested in seeing if it is really worth it to go with the asrock board as a slower upgrade path or if I would have been better served to wait a couple of months to get the scratch for a better board + DDR2.

Any specific types of benchmarks folks would be interested in before I rip my rig apart to build the C2D? (Within reason of course.... I wanna play with the new gear sometime this week, so I won't be entertaining any 2 day prime95 fests..... :) ).

[Shadowed]

Are you going to use the 1950pro on that board ?
It will be interesting to see if there is much noticable difference running it on a 4x pcie bus.

Oh, btw, Happy birthday :)

[sswilson]

Yep, for now I'll be using everything but the 3700 and the DFI board. (My young lad will just have to live with a couple of 512 sticks of ddr400 kingston value ram until I get the money for DDR2 800).

The X4 bus speed caught me by surprise as everybody has it listed as a PCIe 16X motherboard. I'm guessing that it's technically a X16 slot, but it shares the same X4 bus as the AGP slot.

So far, everything I've read leads me to believe that the GPU performance drop will be around 10%, but I'm keeping my fingers crossed that the boost from the dual core CPU will make up the difference in gaming.

I'm going to do the standards (AM2, 3DMark `03/`05/`06) for GPU, and probably try to come up with a file conversion (MP3 --> WMA) test and super-pi for synthetic/real cpu benchmarks.

Does anybody know of a freeware FPS benchmark based on a single game that I could download? (Like the fear or similar tests you see on most reviews).

[Shadowed]

I have heard of people using a utility called fraps. I havn't tried it myself though.

http://www.digital-daily.com/video/vga_testing_2007/

[Babrbarossa]

I have Fraps and have tried it out- I find that it's very inconsistent in it's result output- not through the fault of the program, but the variables involved with the actual games. I think the best thing to do is rely on games that have built-in benchmarks- I read an article to the same effect:

http://www.digit-life.com/articles2/video/fraps.html

[Scratch]

yes Fraps are for games without an integrated bench demo. You can simply run Fraps in the background in the corner of your screen and use a specified area within a game as your benchmark then calculate your "mean" fps so you have a realtime gaming bench
It is also a good program for recording ingame events such as castle seiges and/or pvp events, Epic Boss Raids etc

[sswilson]

Yeah, I'm looking more for something that has a built in benchmark, but I'm not a FPS gamer at all, so the only thing I've got which would even come close would be Oblivion and it doesn't have a benchmarking function. (well not that I know of anywise.....).

I'll just stick with the 3Dmark series then and not worry about the "synthetic test" issues.

[Babrbarossa]

I know COH has a good one- very GPU based though

[sswilson]

I'm going to check through my Maximum PC disks to see if any of the demos have a benchmark function.

[Beermaster]

I always though a good benchmark was 3dmark?