Thermaltake Super Orb
- Thread starter Patgod
- Start date
ocworkbench.com and frostytech.com have both put out reviews that clearly show its performance to be highly lacking.
My own tests, not done on crappy socket-thermistor mb setups, show that it performs worse than the retail T-bird(900mhz and up) heatsink.
Mike
My own tests, not done on crappy socket-thermistor mb setups, show that it performs worse than the retail T-bird(900mhz and up) heatsink.
Mike
If you want better aesthetics, take the Super Orb but if you want a real HSF, then take the ones from GlobalWin, Alpha, Taisol or Kanie.
I dunno about that, I use a SuperOrb with my AMD 1.2 GHz Athlon and my average CPU temp is about 15'C lower than my Coppermine P3s. (35'C vs. 50'C) I know it isn't the most scientific test, but I do think I am getting good results with the Super Orb.
Also, the Super Orb has one thing going for it that no other Socket A HSF does: dual fans. That way, if one fails, your CPU doesn't burn up.
Also, the Super Orb has one thing going for it that no other Socket A HSF does: dual fans. That way, if one fails, your CPU doesn't burn up.
Raaboso,
First off, there is no way a comparable speed t-bird(or in your case, faster t-birds) are cooler than p3s. herein lies hte first problem, socket-a boards do not measure core temp, and some boards do not measure even close to core temp.
Secondly, Socket-thermistor readings from Super/Chrome Orbs are highly suspect, as they do something to effect socket-thermistor readings. One might ask, how are they effected? Very simply, even aiming a can of compressed air at the socket-region for 1 second will drop the reading 4C. Flip it upside down so that condesnation forms, and the reading drops 6C.
Thirdly, it looks like you're using an MSI board. THese boards typically under-read temps 12-15C, if not at times much more.
Lastly, even AMD doesn't suggest using socket-thermsitors to test heatsinks. And there is a reason that hte palamino core will finally have an internal diode.
MIke
First off, there is no way a comparable speed t-bird(or in your case, faster t-birds) are cooler than p3s. herein lies hte first problem, socket-a boards do not measure core temp, and some boards do not measure even close to core temp.
Secondly, Socket-thermistor readings from Super/Chrome Orbs are highly suspect, as they do something to effect socket-thermistor readings. One might ask, how are they effected? Very simply, even aiming a can of compressed air at the socket-region for 1 second will drop the reading 4C. Flip it upside down so that condesnation forms, and the reading drops 6C.
Thirdly, it looks like you're using an MSI board. THese boards typically under-read temps 12-15C, if not at times much more.
Lastly, even AMD doesn't suggest using socket-thermsitors to test heatsinks. And there is a reason that hte palamino core will finally have an internal diode.
MIke
so biggs, if the super orb really just a painting? i mean VR is cool and all, but i would rather my HSF be made out of like...atoms....
Actually, I'm using an A7M266. I do realize that it is hard to take or compare CPU temps, however, I can't believe that it is under-reading the temp by 15'C.
Anyway, you didn't respond to my other point: that the SuperOrb has 2 fans. Even if my CPU is 5 or 10 degrees higher than it would be with a different fan/heatsink, I think it's worth it because of the chance of a fan failure.
--Hans
Anyway, you didn't respond to my other point: that the SuperOrb has 2 fans. Even if my CPU is 5 or 10 degrees higher than it would be with a different fan/heatsink, I think it's worth it because of the chance of a fan failure.
--Hans
No, actually, i would rather have a .35C/W heatsink like a PEP66 that has enough capacity to handle a cpu/fan failure and set my monitoring program to shutdown the system when the fan fails.
That, versus have a .62C/W heatsink, which runs a 1.2ghz t-bird, core temperature wise, some 35-40C over ambient case temp.
Socket-a boards do not read temp well at all, and the problem gets worse with poorer working heatsinks(since compression is non linear, it has a more "adverse" effect on the worse heatsink) or at high wattages, and the situation is at the absolute worse with a sub-par heatsink at high wattages.
What bios is your A7M using?
I've seen Kt7's, like the Anandtech review one, with the UL compensated bios under-read temps by up to 22C.
The fact that your t-bird at 67W is supposedly running cooler than your 30W p3s is one indication that hte temp readings are highly fubared...
Mike
That, versus have a .62C/W heatsink, which runs a 1.2ghz t-bird, core temperature wise, some 35-40C over ambient case temp.
Socket-a boards do not read temp well at all, and the problem gets worse with poorer working heatsinks(since compression is non linear, it has a more "adverse" effect on the worse heatsink) or at high wattages, and the situation is at the absolute worse with a sub-par heatsink at high wattages.
What bios is your A7M using?
I've seen Kt7's, like the Anandtech review one, with the UL compensated bios under-read temps by up to 22C.
The fact that your t-bird at 67W is supposedly running cooler than your 30W p3s is one indication that hte temp readings are highly fubared...
Mike
No, actually, i would rather have a .35C/W heatsink like a PEP66 that has enough capacity to handle a cpu/fan failure and set my monitoring program to shutdown the system when the fan fails.
I guess we have different priorities. I need my computer to continue working at least for a short time after a fan failure.
That, versus have a .62C/W heatsink, which runs a 1.2ghz t-bird, core temperature wise, some 35-40C over ambient case temp.
I doubt it is actually that hot. I have several auxilary fans as well which help the situation...
What bios is your A7M using?
1003. Reported similar temps with 1002.
The fact that your t-bird at 67W is supposedly running cooler than your 30W p3s is one indication that hte temp readings are highly fubared...
I'm not saying it is reading the correct temp, I'm saying it works for me and I'm pleased with the end result - it is stable and it ain't on fire
I guess we have different priorities. I need my computer to continue working at least for a short time after a fan failure.
That, versus have a .62C/W heatsink, which runs a 1.2ghz t-bird, core temperature wise, some 35-40C over ambient case temp.
I doubt it is actually that hot. I have several auxilary fans as well which help the situation...
What bios is your A7M using?
1003. Reported similar temps with 1002.
The fact that your t-bird at 67W is supposedly running cooler than your 30W p3s is one indication that hte temp readings are highly fubared...
I'm not saying it is reading the correct temp, I'm saying it works for me and I'm pleased with the end result - it is stable and it ain't on fire
You really don't think its running that hot? How about this test at frostytech...
FrostyTech TestBed results
@50W, a simulated core temp some 35C over ambient. You're using a 67W processor. At full load, your cpu will be running, core tmep wise, at a minimum 35C over ambient, and probably more like 40-45C over ambient.
Or the ocworkbench review, where, despite using a core-edge thermistor, the 45W t-bird 900mhz with super-orb ran some 22C over ambient.
1.2ghz is the very edge of effectiveness of Super-Orb performance. You won't get much overclocking, and if that's your intention, another heatsink is definately in order.
Heck, there was a bbs member who swithced from a super-orb @1ghz to a fop32-1 and magically, his was now able to overclock his CPU. It isn't magic, is mis-reporting of temps by socket-a boards and poor reviews that claim that hte super-orb is the equivelant of a alpha or globalwin heatsink.
Mike
FrostyTech TestBed results
@50W, a simulated core temp some 35C over ambient. You're using a 67W processor. At full load, your cpu will be running, core tmep wise, at a minimum 35C over ambient, and probably more like 40-45C over ambient.
Or the ocworkbench review, where, despite using a core-edge thermistor, the 45W t-bird 900mhz with super-orb ran some 22C over ambient.
1.2ghz is the very edge of effectiveness of Super-Orb performance. You won't get much overclocking, and if that's your intention, another heatsink is definately in order.
Heck, there was a bbs member who swithced from a super-orb @1ghz to a fop32-1 and magically, his was now able to overclock his CPU. It isn't magic, is mis-reporting of temps by socket-a boards and poor reviews that claim that hte super-orb is the equivelant of a alpha or globalwin heatsink.
Mike
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
-
-
Facebook
Twitter