|
Looking to Build a New Gaming PC
|
|
|
Mackles
Newbie
|
30. December 2008 @ 00:58 |
Link to this message
|
|
This message has been edited since posting. Last time this message was edited on 30. December 2008 @ 01:00
|
|
Advertisement
|
 |
|
|
|
gera229
Member
|
30. December 2008 @ 01:11 |
Link to this message
|
Well the black edition would be better in performance in gaming though. I'm mostly doing this for gaming though. Any my ram says 1066mhz. Well I think the overall performance would be more than enough.
|
|
JaguarGod
Senior Member
|
31. December 2008 @ 03:26 |
Link to this message
|
@Mackles,
Both cases should fit everything inside. The Coolermaster is a larger case, so there would be more room inside and possibly better airflow. It has a bracket for 4 side fans which would mean that it would be cooler inside than since the fans will suck in cooler air from outside the case. Also, this Case is ATX/BTX, so it can be used for BTX motherboard later on. The Antec seems like it will be more silent. The case seems to have a lot of airflow and looks to be a very good case. I am not too keen on cases, so maybe wait for someone who is or do some searches for reviews of cases.
@gera229
The GTX 260 Black Edition is a faster card, but it is still similar in performance. Basically, there would not be a game playable with the GTX 260 that is unplayable with the HD 4870 1GB edition at the same settings and resolution. Each card also has games that it would excel in, so they would trade off victories should they go head to head. Overall though, for the extra $20 or whatever, the GTX 260 Black Edition is definitely a good option. At regular price, it is not as attractive since there is the similar performing HD 4870 1GB at much lower price point and the better performing HD 4850x2 at a similar price point.
Oh, and I have new results for that same 140 minute video. I installed the software to a newer version (actually H.264 v2.1) and time improved to 9 minutes 23 seconds. That is a 10.676:1 ratio, so it encodes almost 11 seconds of mpeg2 video in 1 second!!!
This represents a 22% increase in speed. The CPU was loaded to an average of about 62%, so it really used about 2 1/2 cores :( However this is a 43% increase in core usage and 72% scaling efficiency. Predicted time of a 100% load encode is now 6 minutes 52 seconds.
Theoretical max is 5 minutes 17 seconds
Theoretical Time is 5 minutes 43 seconds
Predicted time is 6 minutes 52 seconds
I think what the Theoretical Max would represent is the difference between C2Q and i7 in encoding. The i7 would probably approach the 5:17 mark. This means that it would be about 23.144% faster. This means that at 3.2GHz a C2Q would be about as fast as an i7 at 2.46GHz.
So here is how the i7's would compare to a 65nm C2Q:
i7 920 = 3.461 GHz
i7 940 = 3.812 GHz
i7 965 = 4.164 GHz
To 45nm C2Q:
i7 920 = 3.331 GHz
i7 940 = 3.647 GHz
i7 965 = 3.983 GHz
So, what these numbers mean is that is you have a QX9650 and were wondering how much you would have to overclock to get it to equal a certain i7 CPU, you would look at the 45nm "chart" and look at the equivalent clock speed. This means that to match an i7 940, a QX9650 would have to be overclocked to 3.467 GHz. Of course this is all theoretical. It would be nice to see how this would compare to actual data.
However, this is NOT gaming performance. This would most likely fit only for very CPU intensive tasks like encoding of audio/video, and rendering of 2D/3D content. It would not make a difference in everyday usage. It is more of a benchmark difference. Actually based on what I have done, there is a very good chance that a C2D at 4GHz would render video faster than a C2Q since the software was not utilizing all 4 cores 100%. H.264 was not cheap. I believe it was $500, so if this software cannot do it, then it is very unlikely that lower end software can.
I will try something with Mainconcept Reference when I get a chance. Also, I will have to test out Adobe Products with the GPU acceleration. I should have a 2600XT somewhere around and that is not a supported card, so it would be a cool benchmark :P
|
|
gera229
Member
|
31. December 2008 @ 15:14 |
Link to this message
|
|
Thanks. Well yea so what do you think would be best for me a C2Q or i7? Basically I usually don't do any video editing. Or w/e. And what else is i7 better than core 2 quad for and by how much better(you can list in percentage % amount better if so)? And in the future for C2Q software will utilize all 4 cores IMO. Oh So if i7 finishes it in 5 minutes the C2Q would finish it in about 6 minutes 30 seconds? Thanks. Peace.
This message has been edited since posting. Last time this message was edited on 31. December 2008 @ 15:32
|
|
JaguarGod
Senior Member
|
31. December 2008 @ 19:00 |
Link to this message
|
For you, I think the Q9550 is the best option. It will cost $380 less than an i7 920 build and will outperform it in gaming.
Also, note that the times I listed were for and i7 clocked at 3.2GHz only if it were utilized 100%, which may not happen outside of 3D animation rendering. For the i7 920, you are looking at times around 6:30 at stock speed, so a little bit faster than the Q6600 @ 3.2GHz.
I think the i7 would be 23% faster at everything clock for clock except for gaming. Now, whatw does 23% mean for everyday computing? Instead of launching FireFox in .1 seconds, .08 seconds? Honestly, you will not see the difference unless you do something where the CPU is utilized at a high percentage over a long period of time.
Only Encoding/Rendering can do this. The only normal utilities that can stress the CPU over some period of time are Anti-Virus software and Compression utilities (7zip, WinrRAR etc...). Antivirus might be more HDD limited since it will access many files. So maybe if you compress and decompress a lot of DVD sized archives, the i7 will make a difference of 20% or so.
I have no idea how long each will take as I did not really think to take note of how long it took to decompress this stuff. I think 1GB took 85 seconds. It would take anywhere from 60 - 70 seconds for an i7 I guess... So if you do this a lot, it would make a difference, but again, 65 vs. 85 seconds for 1GB is not something that would be noticed anyways. This would be a 3 minute difference in a DVD9 sized decompression, or 11:46 vs. 9:00. This is not the end of the world :P
|
|
gera229
Member
|
31. December 2008 @ 19:26 |
Link to this message
|
|
Thanks. In the cyberpower website I see some cheap computers, but I'm not sure weather to build one or get it from them. What will be cheaper? They took a $1700 computer and put all the same exact parts on newegg and on newegg it was $2056 all together. I thought it was supposed to be cheaper to build a pc. Or was there like something they did on purpose like making there be no combo deals? Or what is the problem why is it more expensive? WEll in cyberpower they have liquid cooled pcs and the 1700 one was liquid cooled. Thanks.
This message has been edited since posting. Last time this message was edited on 31. December 2008 @ 19:26
|
|
JaguarGod
Senior Member
|
31. December 2008 @ 20:01 |
Link to this message
|
List the components you are looking at. If it is cheaper from cyberpower, then it doesn't make sense to spend more and build yourself.
You have to watch out with the RAM though. The Cyberpower does not guarantee RAM unless you pay extra.
I don't know much about water cooling. Maybe there is bad/cheap water cooling and good/expensive water cooling.
|
|
Advertisement
|
|
|
|
gera229
Member
|
31. December 2008 @ 20:31 |
Link to this message
|
|
|
