Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
I'm looking to upgrade my 2010 Mac Pro with dual Xeon 5680s and came across a matched pair on eBay.

I asked the seller if they had b1 stepping he looked on the bois and found this 06-2c-2 he only had a server to test so couldn't download CPU-Z or other apps to check.

Is there another way of telling if a CPU has B1 Stepping?
 

Attachments

  • Screenshot_20181122-133537.jpg
    Screenshot_20181122-133537.jpg
    761.3 KB · Views: 757
Last edited:

tsialex

Contributor
Jun 13, 2016
13,450
13,599
I'm looking to upgrade my 2010 Mac Pro with dual Xeon 5680S and came across a matched pair on eBay.

I asked the seller if they had b1 stepping he looked on the bois and found this 06-2c-2 he only had a server to test couldn't download CPU-Z or other apps to check.

Is there another way of telling if a CPU has B1 Stepping?

All Xeons X5680 are B1 stepping, even ES/QC ones. http://www.cpu-world.com/CPUs/Xeon/Intel-Xeon X5680 - AT80614005124AA (BX80614X5680).html
 

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
Thanks for the info. Is this also the case for the x5675 CPUs regarding B1 stepping.
 
Last edited:
  • Like
Reactions: octoviaa

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
Thanks for the help

My setup

8 core 2.4,
Ssd 500gb sdd
2tb hdd,
48gb ram,
evga 980sc

So look forward to the upgrade
 

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
I have just purchased dual X5680s temperature wise what should be a happy range? I have Arctic Silver 5 and Arctic MX-4 which would be the best to use?

Thanks
 
Last edited:

h9826790

macrumors P6
Apr 3, 2014
16,656
8,587
Hong Kong
I have just purchased dual X5680s temperature wise what should range? I have Arctic Silver 5 and Arctic MX-4 which would be the best to use?

Thanks

MX4 should be better.

AS5 may be easier to apply.

Both works good on cMP anyway.

The cMP's native SMC setting may let your CPU warm up to about 80C before spin up the fan, and let it stables at around 85C.

The cooler on cMP is definitely more than adequate (assume the thermal paste is in good condition, and the heatsink is clean, etc). Should be no need to worry about overheat.

However, due to Apple never ship dual 130W CPU configured cMP. Some users worry about the SMC unable to handle the CPU's heat correctly, and manually setup their fan profile. I personally prefer that as well. Even the CPU can work at 85C, but I think still better to give it few more degrees buffer. The native setting is really very close to edge of the thermal envelope. If the heatsink is a bit dusty, or the fan doesn't spin up fast enough etc. The CPU may not able to maintain the Turbo Boost.
 
  • Like
Reactions: alphaod

alphaod

macrumors Core
Feb 9, 2008
22,183
1,245
NYC
I have just purchased dual X5680s temperature wise what should be a happy range? I have Arctic Silver 5 and Arctic MX-4 which would be the best to use?

Thanks

I have dual X5690, last three days I have been running full renders on all the cores, the front CPU is peaking at 70ºC and the rear CPU is topping off at 82ºC.

I have both processors delidded and I'm using one of those IC Graphite thermal pads cut to size.
 

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
MX4 should be better.

AS5 may be easier to apply.

Both works good on cMP anyway.

The cMP's native SMC setting may let your CPU warm up to about 80C before spin up the fan, and let it stables at around 85C.

The cooler on cMP is definitely more than adequate (assume the thermal paste is in good condition, and the heatsink is clean, etc). Should be no need to worry about overheat.

However, due to Apple never ship dual 130W CPU configured cMP. Some users worry about the SMC unable to handle the CPU's heat correctly, and manually setup their fan profile. I personally prefer that as well. Even the CPU can work at 85C, but I think still better to give it few more degrees buffer. The native setting is really very close to edge of the thermal envelope. If the heatsink is a bit dusty, or the fan doesn't spin up fast enough etc. The CPU may not able to maintain the Turbo Boost.


Thanks for the help
Just installed MacFanControl and haven't upgraded the CPUs yet.
What would be the the best settings for all the fans sped etc? I do a lot of GPU and CPU intensive work.

Thanks again.
[doublepost=1543153420][/doublepost]
I have dual X5690, last three days I have been running full renders on all the cores, the front CPU is peaking at 70ºC and the rear CPU is topping off at 82ºC.

I have both processors delidded and I'm using one of those IC Graphite thermal pads cut to size.

I've heard about the Graphic pad but already had paste. How did you get those temperatures?
 

h9826790

macrumors P6
Apr 3, 2014
16,656
8,587
Hong Kong
Thanks for the help
Just installed MacFanControl and haven't upgraded the CPUs yet.
What would be the the best settings for all the fans sped etc? I do a lot of GPU and CPU intensive work.

It depends on what you want.

I personally prefer low fan noise but just give the CPU few more degree (than Apple native setting) buffer from processor hot temperature.

What I am using now is all Intake / Exhaust / CPU Booster fans base on CPU temperature diode temperature.

Min 73C, Max 99C

With AS5, this will keep my CPU below 80C (usually stabilise at 78C) when stressing the CPU to 100% for long time. This is true for up to 36C system ambient (and assume the heatsink / fans are fairly clean)
CPU Temperature.jpg

And now, I replaced AS5 by liquid metal. The CPU can hardly go over 76C with even lower fan speed.

For graphic card, set the PCIe fan speed base on PCIe Ambient temperature.

Min 40, Max 55

This works well for all my 7950, R9 280, R9 380, RX580, 1080Ti, whenever the graphic card warm up the PCIe compartment, the fan will automatically spin up and allow the card's onboard cooler able to do the job properly.
 
  • Like
Reactions: octoviaa

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
It depends on what you want.

I personally prefer low fan noise but just give the CPU few more degree (than Apple native setting) buffer from processor hot temperature.

What I am using now is all Intake / Exhaust / CPU Booster fans base on CPU temperature diode temperature.

Min 73C, Max 99C

With AS5, this will keep my CPU below 80C (usually stabilise at 78C) when stressing the CPU to 100% for long time. This is true for up to 36C system ambient (and assume the heatsink / fans are fairly clean)
View attachment 806436
And now, I replaced AS5 by liquid metal. The CPU can hardly go over 76C with even lower fan speed.

For graphic card, set the PCIe fan speed base on PCIe Ambient temperature.

Min 40, Max 55

This works well for all my 7950, R9 280, R9 380, RX580, 1080Ti, whenever the graphic card warm up the PCIe compartment, the fan will automatically spin up and allow the card's onboard cooler able to do the job properly.

Thanks so much for your help
I think I'll use the MX-4 over AS5, from what I've read that seems to have better durability and temperatures.
Have you use It?

Ideally I want the low noise but want to make sure I get cool temperatures x5680 especially when doing photo/video editing.
I've had my mac for so long, I just don't want it wear out.
The mac is primarily a work machine but I do game as well.

When I do the upgrade I'll follow your temperatures and fan speeds and see how that goes.

Thanks again

Ps. What's the app you're using in pic?
 

tsialex

Contributor
Jun 13, 2016
13,450
13,599
Thanks so much for your help
I think I'll use the MX-4 over AS5, from what I've read that seems to have better durability and temperatures.
Have you use It?

Ideally I want the low noise but want to make sure I get cool temperatures x5680 especially when doing photo/video editing.
I've had my mac for so long, I just don't want it wear out.
The mac is primarily a work machine but I do game as well.

When I do the upgrade I'll follow your temperatures and fan speeds and see how that goes.

Thanks again

Ps. What's the app you're using in pic?
It’s more temperature effective to keep your Mac clean, do periodical cleanups with compressed air and replace the thermal paste every now and then. Mac Pro don’t have CPU thermal problems even on a hot country like mine.

Keep it clean.

App is Bjango iStat Menus.
 

lowendlinux

macrumors 603
Sep 24, 2014
5,460
6,788
Germany
Honestly I just turn the fans up when I'm going to be doing something intense then turn them back down after the job is done.
 

h9826790

macrumors P6
Apr 3, 2014
16,656
8,587
Hong Kong
It’s more temperature effective to keep your Mac clean, do periodical cleanups with compressed air and replace the thermal paste every now and then. Mac Pro don’t have CPU thermal problems even on a hot country like mine.

Keep it clean.

App is Bjango iStat Menus.

Completely agree, keep the heatsink clean is the key.

This chart shows the temperature difference of the North bridge before and after cleaning (same fan speed). Just blow the dust away, nothing really special. And I am not talking about years of dust, but only 2 weeks (oh~ well~ the place I leave is dusty).
NB dust clean.jpg

Once the heatsink is clean, the temperature difference between the CPU diode and system ambient can go as low as 5C on my cMP. But with just a thin layer of dust, this delta T will increase for about 2C, which is 40% in this case actually, quite significant.
5 Delta C.png
 

alphaod

macrumors Core
Feb 9, 2008
22,183
1,245
NYC
Thanks for the help
Just installed MacFanControl and haven't upgraded the CPUs yet.
What would be the the best settings for all the fans sped etc? I do a lot of GPU and CPU intensive work.

Thanks again.
[doublepost=1543153420][/doublepost]

I've heard about the Graphic pad but already had paste. How did you get those temperatures?

I use iStat menus that shows my temperatures.

Interestingly I decided to put liquid metal (Conductonaut) on the Mac Pro yesterday and for some reason it seems my temperatures have gone up…
 

h9826790

macrumors P6
Apr 3, 2014
16,656
8,587
Hong Kong
I use iStat menus that shows my temperatures.

Interestingly I decided to put liquid metal (Conductonaut) on the Mac Pro yesterday and for some reason it seems my temperatures have gone up…

I am also using liquid metal (an unknown brand cheap China stuff), the temperature clearly goes down.

Now even I run Prime95 (small FFT), the CPU diode temperature still can't touch 80C with almost idle fan speed (32C system ambient as per screenshot)
P95 low fan - clean.png


If I allow the fan spin up just a little bit more, the temperature will go down straight away.
P95 standard fan - clean.png


I only use AS5 before liquid metal. So, not really top of the line thermal paste. May be that's why I can clearly see the improvement.

In your case, the graphite thermal pad work better than liquid metal? Did you record data for comparison?

Or how good / bad the cooling performance now?
 

alphaod

macrumors Core
Feb 9, 2008
22,183
1,245
NYC
In your case, the graphite thermal pad work better than liquid metal? Did you record data for comparison?

Or how good / bad the cooling performance now?

Before with my original upgrade, I used Arctic Silver 5, it was hovering around 90ºC on CPU A (rear processor) and 75ºC on CPU B (front processor). I was repasting a couple of my desktops, I got some of those IC Graphite thermal pad that worked well in a couple computers so I put it one in the Mac Pro, seems temps for CPU A dropped to 80ºC and CPU B dropped to around 70ºC. I built a VR gaming PC for my living room and in a HTPC space it was getting very hot, so I did a delid and ran Conductonaut under the IHS and used a thermal pad between the IHS and the heatsink, definitely dropped the temps from 95ºC on load to 80ºC.

I figured on the Mac Pro, I should expect lower if not the same temps, so I took the heatsink off and put liquid metal under there. Anyways it seems my temperatures are back up to 90ºC on CPU A and 75ºC on CPU B which means it's not doing any better than the AS5 I used to have. I may put the thermal pad back in there since it's already cut to size. I tested with Prime95 on max heat as well.

UPDATE: I removed and cleaned up the liquid metal best I could and I put the graphite pads back in there. Now temps are back to 80ºC max on CPU A and 70ºC max on CPU B. This is after 3 hours so it's normalized temperatures. I have a theory perhaps that my delidded CPUs are not fully flat and the copper base is not entirely flat, so when you have a thin coat of liquid metal, there isn't enough contact surfaces whereas the graphite thermal pad is slightly thicker (at 0.1 mm) is just enough for two non-smooth surfaces to work together.
 
Last edited:

h9826790

macrumors P6
Apr 3, 2014
16,656
8,587
Hong Kong
Before with my original upgrade, I used Arctic Silver 5, it was hovering around 90ºC on CPU A (rear processor) and 75ºC on CPU B (front processor). I was repasting a couple of my desktops, I got some of those IC Graphite thermal pad that worked well in a couple computers so I put it one in the Mac Pro, seems temps for CPU A dropped to 80ºC and CPU B dropped to around 70ºC. I built a VR gaming PC for my living room and in a HTPC space it was getting very hot, so I did a delid and ran Conductonaut under the IHS and used a thermal pad between the IHS and the heatsink, definitely dropped the temps from 95ºC on load to 80ºC.

I figured on the Mac Pro, I should expect lower if not the same temps, so I took the heatsink off and put liquid metal under there. Anyways it seems my temperatures are back up to 90ºC on CPU A and 75ºC on CPU B which means it's not doing any better than the AS5 I used to have. I may put the thermal pad back in there since it's already cut to size. I tested with Prime95 on max heat as well.

UPDATE: I removed and cleaned up the liquid metal best I could and I put the graphite pads back in there. Now temps are back to 80ºC max on CPU A and 70ºC max on CPU B. This is after 3 hours so it's normalized temperatures. I have a theory perhaps that my delidded CPUs are not fully flat and the copper base is not entirely flat, so when you have a thin coat of liquid metal, there isn't enough contact surfaces whereas the graphite thermal pad is slightly thicker (at 0.1 mm) is just enough for two non-smooth surfaces to work together.

I am also thinking about "how large the gap on a delidded CPU setup". May be you can tighten the heatsink a little bit more. With delidded CPU, the risk of damaging the socket should be very very low. But if you already removed the liquid metal, I think no point to try that again.

And yeah, 15C off straight away after I delid my 8700K and replace the stock Intel thermal paste by liquid metal. The effect is very noticeable.
 

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
Hi, I bit and did the upgrade yesterday using MX-4 and the line method. I'm concerned about the temps when just booted and in idle.
prochot A is about 60-65c and B is 68-71c
First screenshot is before upgrade and the other two are after. Is this something to worry about? Should I repaste?
Screen Shot 2018-11-28 at 19.19.02 8core.png

Screen Shot 2018-11-29 at 00.59.34.png

Screen Shot 2018-11-29 at 02.11.36 12 core.png
 
Last edited:

h9826790

macrumors P6
Apr 3, 2014
16,656
8,587
Hong Kong
Hi, I bit and did the upgrade yesterday using MX-4 and the line method. I'm concerned about the temps when just booted and in idle.
prochot A is about 60-65c and B is 68-71c
First screenshot is before upgrade and the other two are after. Is this something to worry about? Should I repaste?
View attachment 807248
View attachment 807259
View attachment 807260

You CPU temperatures are very OK. Quite low indeed.

You misinterpret the numbers. That "Temperature to Processor Hot" (ProcHot) is a count down number, the bigger the better. When down to zero, that means you reach the max temperature limit.

If you want to read the temperature in the "traditional" way, you only need to read the CPU Tdiode Temperature. As long as they are below 85C, you are good to go (this is the Apple native setting).

With that temperature, you don't need that high fan speed (in screen capture 2).

And if you compare before and after re-paste. The after one shows better temperature without lower fan speed. Even the system ambient is also lower, but since the Booster fan is 300RPM lower, which means your MX-4 application is good.

Now you can run Prime 95 stress test to make sure it's stable.
 

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
Okay great I'll do the stress test and post the results in a bit jus to add the ProcHot in cpu is now A 50-55c and B is 68. Is this still OK? I might reapply paste probably put slightly to much on.
 
Last edited:

h9826790

macrumors P6
Apr 3, 2014
16,656
8,587
Hong Kong
Okay great I'll do the stress test and post the results in a bit jus to add the ProcHot in cpu is now A 50-55c and B is 68. Is this still OK? I might reapply paste probably put slightly to much on.

As I said, that's a count down counter, 50C from processor hot is more than OK. Do NOT fix anything if they are now working.
 

Mac pro10 5.1

macrumors newbie
Original poster
Nov 22, 2018
17
3
As I said, that's a count down counter, 50C from processor hot is more than OK. Do NOT fix anything if they are now working.
Ok it just that goes down to 35 when editing then back to 50 or there abouts at idle from what you said. I just don't want it to go to low I'll post a screen shot when I'm at the computer.
 

h9826790

macrumors P6
Apr 3, 2014
16,656
8,587
Hong Kong
Ok it just that goes down to 35 when editing then back to 50 or there abouts at idle from what you said. I just don't want it to go to low I'll post a screen shot when I'm at the computer.

IMO, there is nothing to worry about unless it drops to below 10C. You are far from there.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.