I do also have the Evo Express with another NVMe in it but my Evo plus is in a Mac Pro 5’1 and in slot 3 which is a 4x pcie 2.0, that’s the bottleneck. My rx6800xt eats into my second 16x slot, but honestly it’s plenty fast enough at 1500MBs.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Samsung Nvme and super slow Boot on Monterey
- Thread starter Buntschwalbe
- Start date
- Sort by reaction score
Hi everyone! I've been following this thread for a long time now and finally, I accidentally fixed my long-term issue of slow booting MacOS on a Samsung X5 external drive. What I did is upgraded from Monterey to Ventura Beta 10. Cloned my X5 to my T7, Formatted X5, Cloned T7 back to X5, and voila! It boots super fast again. Also I didn't have to disable TRIM. I'm very happy and I hope others will benefit from my experience.
So, what's the remedy? The Ventura update or the formatting/cloning? When you install the Ventura, did it start working faster, or only after formatting and cloning the drive with Ventura installed already on it? Thanks
It's an old post but ... any progress since then? My boot time is around 3min. I can live with it but I think it kind of buffers when I use big images or on video edits. My wife's Mac is on a normal SSD and it runs smoother with benchmark supposed to be 5 times slower.
Man… I’ve got this issue with a gigabyte ssd nvme about 5 minutes to boot (haven’t timed it admittedly) but it’s a 2015 MacBook Pro so no Ventura for me :$
A bit off-topic, but check out OpenCore Legacy Patcher. Ventura running nicely on Early 2015 MBP here.
On-topic, I’ve set ApfsTrimTimeout to 0 for now (non-updatable Samsung 970 EVO Plus 512) and it improved boot times on Ventura by a whole lot. I guess time will tell if there are repercussions from the change but considering the buggy TRIM support this drive has, I’m optimistic it will not be any worse.
Edit: PS. OpenCore/Ventura update made boot times worse, and with ApfsTrimTimeout = 0 they're better than previously, YMMV but there might not be any magical fixes in Ventura.
Last edited:
I updated my X5 to Ventura hoping I wouldn't have to reinstall but unfortunately its still as slow as it was before.
Will report back when I have time to format and reinstall.
Will report back when I have time to format and reinstall.
Turns out nothing wanted to work so I bit the bullet and gin a Samsung 970 Evo Plus, works like a charm and got a lovely speed boost on top
Obviously check your firmware version first, it is possible to update on a mac by the way
Hi Woefi,
Would to tell me how the Samsung Evo Pro is performing now please. I have an iMac 19.1 and am considering the EVO Plus as it is at a significantly reduced price. That said, earlier MacRumours reports of slow boot-up due to TRIM and difficulties with driver installation make me very cautious. As does the Phoenix vs Elpis controller question.
So what speeds have you achieved via Black Magic? - and is your bootup time constant or is it growing longer over the last eight months or more due to TRIM checks?.
Your help would be appreciated.
Australopithicus
PS. This Terminal method of driver installation should help overcome many issues:
If you are still not up to date with Samsung driver this may help:
Australopithicus
As I wrote, my model is the 970 EVO plus, it's the (newer) Elpis variant which had the 3B2QEXM7 firmware on it. I updated it to 4B2QEXM7 via the Samsung Magician using an external windows boot drive.
The boot times are roughly 1 min. But I only reboot once a month because sleep works perfectly, so no "real" problem for me. Just a little bit annoying...
Performance of SSDs depends on the amount of data you transfer in one go. SSDs have an portion of faster (or cached) space and if this fills up then the speed drops to a secondary level.
For this reason I mostly use "AJA System Test Lite" which gives me a gradual curve of what the performance was at a given data amount.
Here the write speed starts at 2800MB/s and drops to 1100MB/s at the 5GB mark for the rest of the test:
Thank you Woefi. Most helpful. I have adopted AJA as a speed guide as a result of your introduction. Being based upon video, it probably represents working load more realistically than than the faster reading of Black Magic. I appreciate your comments.
I asked my original question to see if TRIM had slowed startup as others had complained. I assume you find the 1 minute boot-up unchanged given it is infrequently used.
The drop off in Writes is quite dramatic. Perhaps the Phoenix controller would have had a less severe decline.
Using AJA, I get writes of 693 with Reads of 2585 with the current standard Apple Fusion 128 NVMe with 2TB HDD. Naturally I seek higher Write performance and would expect to get at least 3x improvement if the upgrade is to be worthwhile. I was considering disconnecting the HDD entirely so I get best speed and adaquate internal storage with a 2TB NVMe .
Are you set up as a Fusion Drive or do you run separate drives?
Have you ever speed tested and compared Fusion vs Solo drives?
No FusionDrive. I swapped the original 256gb Apple NVMe for the Samsung 1tb. Then I ordered the missing SSD cable for the empty slot (where your HDD is) and put an 2tb crucial MX500 in. It maxes out at 400-500 mb/s but I use this space as archive/backup where I usually don’t need the speed. Still waaayy better than having a spinning drive spin up from time to time and dragging the whole system down…
Very interesting. You are set up as I was considering. I may be mistaken but I assumed the Fusion system would slow the NVMe SSD to some extent to serve the slower drive. Individual drives solves that possibility. Explain to me about the missing cable please. The HDD is SATA and you purchase a SATA SSD. I have not done the research but I assumed one plugged in the existing HDD SATA cable to the SSD and that was it. What should I purchase?
Last edited:
If you already have a fusion drive then you got all you need.
My configuration only had the apple nvme so the space for the hdd part was empty.
OK. Understood.
Thank you for all the information Woefi. You have been most helpful. I shall hold off any NVMe purchase until the new year and see what deals are offered.
Best Wishes,
Australopithicus
Hi Pepsilondon,
I cannot comment on the X5 but can say I have a Samsung T7 external from which I simply tossed all the accompanying Samsung exe files, re-formatted T7 as APFS with disk utility, then installed Big Sur from Recovery so it can be used as a boot drive. Now using Monterey. All smooth as silk. All recent o/s run on it just as you would expect.
I have been running for over 12 months. Surely TRIM issues only relate to internal NMVE?
Avoid Samsung software if possible and life is easier!
Given the many issues with Samsung drives, I am not prepared to risk purchase.
I would appreciate readers installation experiences with the
I appreciate my iMac 19.1 is PCIE 3.0, but I get the impression from reviews that a PCIE 4 drive will give higher PCIE 3 speeds.
All comments welcome.
I would appreciate readers installation experiences with the
ADATA XPG Gammix S50-Lite 1TB PCIe Gen4x4 M.2 2280 SSD
Does this NVME install happily in iMac 19.1?I appreciate my iMac 19.1 is PCIE 3.0, but I get the impression from reviews that a PCIE 4 drive will give higher PCIE 3 speeds.
All comments welcome.
Hi MowgliWolf,Turns out nothing wanted to work so I bit the bullet and gin a Samsung 970 Evo Plus, works like a charm and got a lovely speed boost on top
Obviously check your firmware version first, it is possible to update on a mac by the way
Now a few months have past how are your iMac start-up times performing? Are they constant or getting longer?
Which Controller do you have on the Evo Plus; Phoenix or Elpis?
PS. Regarding your external drive start-up issues you commented on: My external Samsung T7 has worked like a charm. I simply trashed the Samsung software, connected, formatted and all has been 100% for over a year.
For the internal Samsung 970 Evo Plus, several reports confirm installing the Samsung ISO on a USB is the most reliable installation method for any mac o/s. I also found this method using Terminal most instructive:
Give me your feedback on SSD performance, meanwhile I hope this info helps fix your problems.
Cheers.
Last edited:
OCLP 0.6.x finally added the ability to control this through the GUI.
- Add APFS Trim Configuration
- Settings -> Misc Settings -> APFS Trim
Switching Trim off altogether reduces the efficiency of your SSD and increases wear on the NAND cells. thus shortening SSD life while making the SSD process slower as NAND has to self clean blocks.
Thanks to Backslashnl1, for the following article from TonyMac86 which explains the issue all mac users face with Samsung:
EXPLANATION]
The APFS filesystem is designed in a way that the space controlled via the spaceman structure is either used or free. This may be different in other filesystems where the areas can be marked as used, free, and unmapped. All free space is trimmed (unmapped/deallocated) at macOS startup. The trimming procedure for NVMe drives happens in LBA ranges due to the nature of the DSM command with up to 256 ranges per command. The more fragmented the memory on the drive is, the more commands are necessary to trim all the free space. Depending on the SSD controller and the level of drive fragmenation, the trim procedure may take a considerable amount of time, causing noticeable boot slowdown. The APFS driver explicitly ignores previously unmapped areas and repeatedly trims them on boot. To mitigate against such boot slowdowns, the macOS driver introduced a timeout (9.999999 seconds) that stops the trim operation when not finished in time. On several controllers, such as Samsung, where the deallocation process is relatively slow, this timeout can be reached very quickly. Essentially, it means that the level of fragmentation is high, thus macOS will attempt to trim the same lower blocks that have previously been deallocated, but never have enough time to deallocate higher blocks. The outcome is that trimming on such SSDs will be non-functional soon after installation, resulting in additional wear on the flash.
One way to workaround the problem is to increase the timeout to an extremely high value, which at the cost of slow boot times (extra minutes) will ensure that all the blocks are trimmed. Set this option to a high value, such as 4294967295, to ensure that all blocks are trimmed. Alternatively, use over-provisioning, if supported, or create a dedicated unmapped partition where the reserve blocks can be found by the controller. Conversely, the trim operation can be disabled by setting a very low timeout value. e.g. 999. Refer to this for details.
[REQ TRANSLATION] url: Можно ли эффективно использовать SSD без поддержки TRIM?
Thanks to Backslashnl1, for the following article from TonyMac86 which explains the issue all mac users face with Samsung:
EXPLANATION]
The APFS filesystem is designed in a way that the space controlled via the spaceman structure is either used or free. This may be different in other filesystems where the areas can be marked as used, free, and unmapped. All free space is trimmed (unmapped/deallocated) at macOS startup. The trimming procedure for NVMe drives happens in LBA ranges due to the nature of the DSM command with up to 256 ranges per command. The more fragmented the memory on the drive is, the more commands are necessary to trim all the free space. Depending on the SSD controller and the level of drive fragmenation, the trim procedure may take a considerable amount of time, causing noticeable boot slowdown. The APFS driver explicitly ignores previously unmapped areas and repeatedly trims them on boot. To mitigate against such boot slowdowns, the macOS driver introduced a timeout (9.999999 seconds) that stops the trim operation when not finished in time. On several controllers, such as Samsung, where the deallocation process is relatively slow, this timeout can be reached very quickly. Essentially, it means that the level of fragmentation is high, thus macOS will attempt to trim the same lower blocks that have previously been deallocated, but never have enough time to deallocate higher blocks. The outcome is that trimming on such SSDs will be non-functional soon after installation, resulting in additional wear on the flash.
One way to workaround the problem is to increase the timeout to an extremely high value, which at the cost of slow boot times (extra minutes) will ensure that all the blocks are trimmed. Set this option to a high value, such as 4294967295, to ensure that all blocks are trimmed. Alternatively, use over-provisioning, if supported, or create a dedicated unmapped partition where the reserve blocks can be found by the controller. Conversely, the trim operation can be disabled by setting a very low timeout value. e.g. 999. Refer to this for details.
[REQ TRANSLATION] url: Можно ли эффективно использовать SSD без поддержки TRIM?
I suspect this issue is not related to Monterey, but has something to do with macos firmware, which is updated very often when a new macos version is installed.
I have an imac 27'' 2015 wich originally has hard drive. Changed then with 970 evo and had no issues until upgraded from big sur to monterey. In monterey it started the slow boot issue.
So I made e fresh new Catalina install. Fine at the beginning but now it start slowing again.
Just put in terminal the command to get trim times on boot (log show --last boot | grep "trims took") day after day, and it is very clear that trim time is increasing again.
Looking at this site you get the new firmware version that is upgraded along with macos new release and actually the firmware for my imac changed twice from catalina to a middle big sur version and then from big sure to monterey.
eclecticlight.co
Sadly, if confirmed, this is a very bad news, since you can downgrade the macos system, but the firmware version remains the latest.....
I have an imac 27'' 2015 wich originally has hard drive. Changed then with 970 evo and had no issues until upgraded from big sur to monterey. In monterey it started the slow boot issue.
So I made e fresh new Catalina install. Fine at the beginning but now it start slowing again.
Just put in terminal the command to get trim times on boot (log show --last boot | grep "trims took") day after day, and it is very clear that trim time is increasing again.
Looking at this site you get the new firmware version that is upgraded along with macos new release and actually the firmware for my imac changed twice from catalina to a middle big sur version and then from big sure to monterey.
Which firmware should your Mac be using? (version 6)
This article lists the firmware versions of Macs which have been successfully updated to run macOS 12.7.4. Apple doesn’t provide an official list of the current firmware versions which should…
eclecticlight.co
Sadly, if confirmed, this is a very bad news, since you can downgrade the macos system, but the firmware version remains the latest.....