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

Highpoint 7101A - PCIe 3.0 SSD performance for the cMP

Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.
dorianclive said:
Yea I’ve been looking for a lid. Perhaps I may need to go back to stock and deal with the loud fan. *sigh*
Click to expand...

Remove the heatsinks from your blades and put the shroud back on with the replacement fan.

Can you share a close up photo of the PEX switch surface?

The sticky thermal surface on the 17mm heat sink I am using (recommended in my previous post) is strong and durable so I am sure you won't have any issues attaching it to the switch surface.

Edited: corrected as I thought you were referring to the card's fan speed.
 
w1z said:
Remove the heatsinks from your blades and put the shroud back on with the replacement fan.

Can you share a close up photo of the PEX switch surface?

The sticky thermal surface on the 17mm heat sink I am using (recommended in my previous post) is strong and durable so I am sure you won't have any issues attaching it to the switch surface.

Edited: corrected as I thought you were referring to the card's fan speed.
Click to expand...

3fac1935fa259034c385f79495660c0b.jpg

47c21b8f60a44f2f840e9878d0210a21.jpg

9a3947f4c8058dbad4e8a3de843286d6.jpg

fdcba96b6e47a87752dcce5498e78f03.jpg
 
@dorianclive I don't foresee a problem with using the recommended 17mm heatsink on that die. Just make sure to clean and swab it with an alcohol pad before attaching the heatsink.

Another option you can look at are clip-on heatsinks from Enzo Technologies.

For now, and until you decide on what route you want to take, continue using the card with the shroud and replacement fan.
 
  • Like
Reactions: Chips Stephens and dorianclive
w1z said:
@dorianclive I don't foresee a problem with using the recommended 17mm heatsink on that die. Just make sure to clean and swab it with an alcohol pad before attaching the heatsink.

Another option you can look at are clip-on heatsinks from Enzo Technologies.

For now, and until you decide on what route you want to take, continue using the card with the shroud and replacement fan.
Click to expand...

I’ve heard of Enzo. I gotta check them out.

Thank you for the suggestion. How’s yours been holding up?
 
  • Like
Reactions: Executor
dorianclive said:
How’s yours been holding up?
Click to expand...

Perfectamundo. I'm seeing the same temperatures as I have reported in my mod posts for both blades and PEX switch with PCIe fan auto spinning at 800RPM during the summer months.
 
w1z said:
Perfectamundo. I'm seeing the same temperatures as I have reported in my mod posts for both blades and PEX switch with PCIe fan auto spinning at 800RPM during the summer months.
Click to expand...

That’s amazing. Once I figure out what’s what I’ll PM you. Don’t wanna be that guy derailing the thread or bringing it off topic.

Thanks again
 
  • Like
Reactions: Executor and w1z
dorianclive said:
Spec of dust. Definitely not damaged.

30b6f08ba26799a874537e2031f1c629.jpg
Click to expand...
The die is too small to attach the heatsink with just the heatsink adhesive, but you can try a shim, like some NVIDIA GPUs with exposed dye have, or even a foam spacer like old Athlon XP/Duron had, and glue the heatsink with 1/3 long cure epoxy and 2/3 white silicone thermal paste. It's a very common formula to attach heatsinks to exposed dyes like this one. I'd only do that after warranty expired.
 
tsialex said:
The die is too small to attach the heatsink with just the heatsink adhesive, but you can try a shim, like some NVIDIA GPUs with exposed dye have, or even a foam spacer like old Athlon XP/Duron had, and glue the heatsink with 1/3 long cure epoxy and 2/3 white silicone thermal paste. It's a very common formula to attach heatsinks to exposed dyes like this one. I'd only do that after warranty expired.
Click to expand...

Good call. I’ll keep that in mind. FWIW I bought the card used off eBay.
 
I have a couple of questions after reading through the thread. I bought my card through a Amazon Warehouse deal and was fortunate to get one that the fan control was optional. (If not I really cannot handle the shrill sound of the stock fan)

I am running the 7101a in slot one of my cMP Mid 2010 and have the fan controller set to the off. The temperature threshold is set to 140 but the two WD Sn750 1Tb drives never seem to get above 130º even when running the Blackmagic Test for extended time. I have the two SSDs configured in the HighPoint RAID and am waiting for a third card to complete my setup.

1) I know from looking at the HighPoint site that the Management software will cause an alert if the monitored temperature goes about the set 140º but will it also trigger the cards fan to spin up in order to drop the temp?

2) Running in stock configuration with fan off and drives remaining cool - do I need to be concerned that the PEX switch will overheat - is there a way to monitor/see the temp of the PEX in the HighPoint software?

3) I am trying to determine if an additional WD Sn750 1Tb drive will be sufficient to set up as my boot drive with user accounts located on the 2Tb RAID. Everything currently exists on a single 2Tb SSD with 500Gb available. How should I go about splitting the data and system between a new NVMe boot drive and the RAID array? (Migration Assistant / Super Duper / ???)

Any suggestions and concerns would be appreciated :)
 
Last edited:
Hey everyone. I finally have 4 970 Pro 512GB in my Highpoint card. I ran Black Magic this morning to check my speeds and admittedly was expecting much faster results.

My setup is as follows:

1st 970 runs the OS and the last 3 I use in a RAID 0 config I created using Apple's RAID Assistant.

Am I doing something wrong here? I was expecting a much faster result. Any help is appreciated. I've attached a screenshot below.

DiskSpeedTest.png
 
^^^^If they are not RAIDED your numbers are fine.

My 2019 NcMP and a Highpoint 7103 with a Samsung 970 pro:

TinyGrab Screen Shot 7-13-20, 9.25.01 AM.png


Lou
 
It’s in Slot 1 right now which is why I’m super confused as to the low speeds.

I’m running a Mac Pro 5,1 on Boot ROM 144, Mojave installed if that helps.
 
dorianclive said:
It’s in Slot 1 right now which is why I’m super confused as to the low speeds.

I’m running a Mac Pro 5,1 on Boot ROM 144, Mojave installed if that helps.
Click to expand...
Did you changed the BlackMagic to the correct mount point of the array? Your results are what is expected for an already used/full of data single 970PRO.
 
dorianclive said:
Yea I selected the array as target drive in Black Magic - not sure what the issue is now
Click to expand...
Check PCIe link width and link rate in System Information.app.

For a complete picture, install pciutils and post result of pcitree.sh.

It's not likely that they would be incorrectly set to anything other than PCIe 2.0 x16 for upstream and PCIe 3.0 x4 for downstream but you should rule that out anyway.
 
joevt said:
Check PCIe link width and link rate in System Information.app.

For a complete picture, install pciutils and post result of pcitree.sh.

It's not likely that they would be incorrectly set to anything other than PCIe 2.0 x16 for upstream and PCIe 3.0 x4 for downstream but you should rule that out anyway.
Click to expand...

That may be the issue. This is what I’m seeing for each controller.
0ecff6a4cd005990e8132089ed489419.jpg
 
dorianclive said:
That may be the issue. This is what I’m seeing for each controller.
0ecff6a4cd005990e8132089ed489419.jpg
Click to expand...
Nothing out of normal here.

All M.2 devices are x4 or less and 8.0GT/s link speed is the internal PCIe v3.0 connection from the PCIe switch to the M.2 device.
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back