Mushkin Rev.3 Buyers, Read This Before You Shop,hehehe

[[OC]_SR20DE]

Please take your time and try to enjoy reading this slowly or you'll get bored like hell.

I received a 128mb Mushkin PC133 HighPerformance Rev.3 today. My plan was to add this new RAM to my existing PC150 HSDRAM 128mb. ** I shouldn't have bought PC133 Rev.3**in my situation. I'll tell you why......


Before we go into some Quaky 3A FPS benchy testing, be aware the letter " H "- "High speed" in HSDRAM of PC150 module. It has its meaning, it's not just there to appeal to our eyes. We know that the PC150 runs @ cas 322. and PC133 Rev.3 runs @ 222. I know that the PC 150 runs access time of 4.5ns but the Rev3 is quite slower. Now,.. here's the benchmark with different RAM and CAS setting.. Note!!!! cas setting 322 is just the way it's arranged Top to Bottom role in my mobo bios but its same damn thing as 232 so don't get confused.


Note: all these tests were run repeatedly at different time intervals, i.e.. re-runs for 3 times and restarting the PC and run 3 times in Quaky 3A TimeDemo 001.


Quake 3A TimeDemo 001/ 1024x768x16bit/ medium high level of Eye Candy on with Default textures, High detail, Dynamic Lights On.. and according to my system setup of Win98SE, P3@ 1064mhz / Annihilator2 GTS 32mb@ 220/365 w/ Good'ol Fastest FPS generating Det. 6.18 driver:


128mb of PC150 HSDRAM @cas 322: blazing 9.4sec/ 144.1 fps

128mb of PC133 Rev.3 @ cas 322: 9.5sec/ 140.6 fps

128mb of PC133 Rev.3 @ cas 222: 9.4sec/ 144.3 fps

256mb of PC133& PC150 @cas 322: 9.4sec/ 142.5 fps


Conclusion: Notice that the PC150 hsdram wipes out PC133 Rev.3@ 322 by 3.5 sec and 0.1 sec quicker completing the demo run time.
The FPS difference betw. PC150 hsdram@ cas 322 and PC133 Rev.3 @ cas 222is only 0.2 sec. That is almost no difference.

Sucky fps performance of Rev.3@ cas322.. this thing can't compare to the PC150's fps performance. You realize that PC150's cas 322's fps performance is virtually the same as PC133 Rev3@ cas222. And horrible fps performance with having both sticks of RAM mixed.. euckk!!.......

It is true that you'll get little higher memory bandwidth benchmark score with a PC133 rev.3@ cas222 than PC150 hsdram in Sandra testing but that's just the memory part.. Look at more of useful apps like gaming fps.. you start to see and more appreciation of the performace of PC150 hsdram. Its clock cycle and access time of 5.6 ns and 4.5 ns does its job. Don't only depend on CAS222. Based on my observation and experience, I'm credible enough to share this information with ya. This might help others who seek RAM modules.

[[OC]_SR20DE]

Oh by the way, Mushkin was so kind enough to exchange the RAM to PC150 hsdram and they just now sent the stick of PC150 hsdram 128mb today before I get to ship this Rev3 module back to them. How nice...

[marty]

Thanks for posting this. I think I have only two questions. One I believe you answered. Is the PC 150 RAM that you have also from Mushkin?

The second question concerns your times that you mention in the post. You posted:

128mb of PC150 HSDRAM @cas 322: blazing 9.4sec/ 144.1 fps

128mb of PC133 Rev.3 @ cas 322: 9.5sec/ 140.6 fps

128mb of PC133 Rev.3 @ cas 222: 9.4sec/ 144.3 fps

256mb of PC133& PC150 @cas 322: 9.4sec/ 142.5 fps


Conclusion: Notice that the PC150 hsdram wipes out PC133 Rev.3@ 322 by 3.5 sec and 0.1 sec quicker completing the demo run time.
The FPS difference betw. PC150 hsdram@ cas 322 and PC133 Rev.3 @ cas 222is only 0.2 sec. That is almost no difference.


But I don't see any times that differ by 3.5 seconds. Am I missing something or is there a typo?

Thanks.

[[OC]_SR20DE]

marty (Mar 14, 2001 04:17 p.m.):
Thanks for posting this. I think I have only two questions. One I believe you answered. Is the PC 150 RAM that you have also from Mushkin?

The second question concerns your times that you mention in the post. You posted:

128mb of PC150 HSDRAM @cas 322: blazing 9.4sec/ 144.1 fps

128mb of PC133 Rev.3 @ cas 322: 9.5sec/ 140.6 fps

128mb of PC133 Rev.3 @ cas 222: 9.4sec/ 144.3 fps

256mb of PC133& PC150 @cas 322: 9.4sec/ 142.5 fps


Conclusion: Notice that the PC150 hsdram wipes out PC133 Rev.3@ 322 by 3.5 sec and 0.1 sec quicker completing the demo run time.
The FPS difference betw. PC150 hsdram@ cas 322 and PC133 Rev.3 @ cas 222is only 0.2 sec. That is almost no difference.


But I don't see any times that differ by 3.5 seconds. Am I missing something or is there a typo?

Thanks.


Sorry about that.. I thought I mentioned what brand my RAM was.. It is a Mushkin PC150 hsdram. It also indicated in my signature..


About the deal with 3.5 sec difference, the information was not super clear.. but it's in there.. if you look at...

Quote "Conclusion: Notice that the PC150 hsdram wipes out PC133 Rev.3@ 322 by 3.5 sec and 0.1 sec quicker completing the demo run time." OOPS.... so sorry.. it's 3.5 FPS, not sec.. sorry.. I typed too much there and so easy to make mistakes.. doh!! lol Thanks for pointing this out, you corrected me..

[DocClock aka MadClocker]

First off, let me remind you that not all mobo's can run hsdram...even with a bios update,
And second, those are horrible scores..although I can't brag either...my freind has a PIII running 1050, and he gets those scores(144fps or so) with an old tnt2 ultra.
If you want better scores, try dumping the opengl dll in your system, and use the 508 opengl.dll with the 618 set, and I'll bet your scores go up.
I am using the 717driverset with the 508 opengl.dll, and with my sucky card not clocked,I get consistant 86fps...but this card does NOT scale with the cpu like the tnt2 Ultra does. @ 933, 92 opengl mips, @ 1059 92 opengl mips.
You do make a good point though...I bought some HSdrsm from Mushkin, and it wouldn't do over 124mhz, so I exchanged it for some rev2, and it will run154mhz cas2-2-2 5/7....and so far, I have had the rev3 up to 156cas2-2-2-5/7, and both sticks will do 158cas3...I have not tested it further. I also thought the hs sticks were a little tall as well.

[[OC]_SR20DE]

hmm... I was actually referring to the fps comparison between Muskin's PC133 Rev3 and PC150 hsdram at same overclocked mhz system. But you don't seem to get my drift.... It's not about who gets higher fps by altering the openGL.dll files, that's old...... I am not interested in that and that's not what I was talking about. Please pay attention as to what direction I was going at when you read threads like this, okie dok?? Another important reason for reading is to understand what concept/idea is behind this as a whole.. LOL

[DaNut]

MilkPowder (Mar 14, 2001 07:43 p.m.):
hmm... I was actually referring to the fps comparison between Muskin's PC133 Rev3 and PC150 hsdram at same overclocked mhz system. But you don't seem to get my drift.... It's not about who gets higher fps by altering the openGL.dll files, that's old...... I am not interested in that and that's not what I was talking about. Please pay attention as to what direction I was going at when you read threads like this, okie dok?? Another important reason for reading is to understand what concept/idea is behind this as a whole.. LOL

I think Doc simply trying to point out the fact that some motherboards are not compatible with Mushkin PC150 HSDRAM, while Mushkin Rev. 3 modules (from our experience) has better compatibility amongst most if not all current generation of motherboards.

[warduck]

I thought id let u know that your test is totally invalid. Infering mainboard ramspeed from vidtesting is about as valid as determining vidio speed from a hard drive speed test.
Note, im being very nice.

If yu would like to know how to test mainboard ramspeed just ask.

warduck