Question Should TRIM be on or off for an NVMe SSD on an iMac ?

[Australopithicus]

I am upgrading an Intel iMac model 19.2 running Ventura, using PCIE NVME Adata SX8600.
I would like to know if this NVME controller, SM2262ENG BC, operates its own internal garbage control system. If it does, I need to switch TRIM off in Ventura as it will reduce SSD performance.
I cannot get any support from Adata website and hope there is a Guru who can help me with this question.
Many Thanks in advance.

 

[CountMike]

I am upgrading an Intel iMac model 19.2 running Ventura, using PCIE NVME Adata SX8600.
I would like to know if this NVME controller, SM2262ENG BC, operates its own internal garbage control system. If it does, I need to switch TRIM off in Ventura as it will reduce SSD performance.
I cannot get any support from Adata website and hope there is a Guru who can help me with this question.
Many Thanks in advance.

There are at least 2 usual ways to invoke TRIM command, from Controller on the SSD and from OS (automatically at preset intervals or manually), in either case it's invoked only when idle and shouldn't negatively impact performance

 

[hotaru.hino]

If an SSD has it's own housecleaning thing, then it's likely any TRIM command just forces it to run. However, I can't imagine why it would reduce performance as TRIM isn't run very often to begin with (Windows does it every week, I'd imagine it's the same for macOS).

 

[Australopithicus]

There are at least 2 usual ways to invoke TRIM command, from Controller on the SSD and from OS (automatically at preset intervals or manually), in either case it's invoked only when idle and shouldn't negatively impact performance

Thank you Mike.
My question was, whether the Adata SSD needed TRIM to be on or off. I understand for Mac O/S, default is OFF because the Apple SSD is tuned to the o/s system and garbage collection is thus integrated with the o/s.
As many recent SSD designs manage their own garbage collection Mac O/S TRIM should be OFF so that it does not reduce SSD performance with conflicting instructions.
Adata are no help. I am inclined to assume the SSD is better without Apple o/s TRim being ON. I would like to know for sure.

 

[Australopithicus]

If an SSD has it's own housecleaning thing, then it's likely any TRIM command just forces it to run. However, I can't imagine why it would reduce performance as TRIM isn't run very often to begin with (Windows does it every week, I'd imagine it's the same for macOS).

 

[CountMike]

Thank you Mike.
My question was, whether the Adata SSD needed TRIM to be on or off. I understand for Mac O/S, default is OFF because the Apple SSD is tuned to the o/s system and garbage collection is thus integrated with the o/s.
As many recent SSD designs manage their own garbage collection Mac O/S TRIM should be OFF so that it does not reduce SSD performance with conflicting instructions.
Adata are no help. I am inclined to assume the SSD is better without Apple o/s TRim being ON. I would like to know for sure.

Sorry but I can't vouch for that particular SSD but looking at Crystal Disk Info shows TRIM enabled on all 5 of my SSDs even if was disabled in Windows.

 

[Australopithicus]

Understood. I think I shall assume the NVME SSD will run better with O/S TRIM off.
Thank you for your input.
Regards,

 

[Australopithicus]

Sorry but I can't vouch for that particular SSD but looking at Crystal Disk Info shows TRIM enabled on all 5 of my SSDs even if was disabled in Windows.

OK Mike. Windows and Macs o/s are very different with TRIM implementation. Recent Mac o/s all have trim OFF by default. From your comment, it looks like it is a standard feature these days to have self supporting Garbage cycle on every SSD.
I think I will keep Ventura o/s TRIM off as I understand it can add wear (conflict), with a self-sufficient SSD garbage system.

Thank you for your input. I reckon it answers the question.

 

[CountMike]

OK Mike. Windows and Macs o/s are very different with TRIM implemetation. Recent Mac o/s all have trim OFF by default. From your comment, it looks like it is a standard feature these days to have self supporting Garbage cycle on every SSD.
I think I will keep Ventura o/s TRIM off as I understand it can add wear (conflict), with a self-sufficient SSD garbage system.

Thank you for your input. I reckon it answers the question.

Yes, but I must guess that modern SSDs incorporate all functions in their controllers, Trim, GC even space allocated for Wear Leveling without having to force it externally. One would have to dissemble firmware for each one to find out about all features,

 

[Australopithicus]

Yes, but I must guess that modern SSDs incorporate all functions in their controllers, Trim, GC even space allocated for Wear Leveling without having to force it externally. One would have to dissemble firmware for each one to find out about all features,

Sure. That is why it is bloody daft not to list the trim feature in product specs.
Apple are just as obstructive. As I said, TRIM is OFF by default. When you enter Terminal to enable Trimforce, you are faced with a capital letter warning that enabling TRIM for third party SSD may result in data loss.

We both know that is typical Apple hostility directed against using non Apple products.

 

[fzabkar]

This question makes no sense. No operating system can possibly know what is going on inside the SSD at the firmware level. Garbage collection is something that the OS cannot see. All the OS can do is to send a standard ATA TRIM command to inform the SSD that particular LBA ranges are free and can be repurposed by the firmware.

Can you show us any URL where Apple claims to have special interactions with the SSD firmware?

BTW, garbage collection happens whether or not TRIM is enabled. It is a necessary consequence of wear levelling. Disabling TRIM just makes garbage collection less efficient.

My advice would be to enable TRIM and enjoy a longer life for your SSD, and better performance. Forget about this Apple/Ventura/TRIM nonsense.

 

[CountMike]

Sure. That is why it is bloody daft not to list the trim feature in product specs.
Apple are just as obstructive. As I said, TRIM is OFF by default. When you enter Terminal to enable Trimforce, you are faced with a capital letter warning that enabling TRIM for third party SSD may result in data loss.

We both know that is typical Apple hostility directed against using non Apple products.

Not surprised, exclusivity with HW, OS and SW is what made Apple stand off in first place, otherwise they would be just another IBM, Dell, Hp... .. which also tried but mostly failed to have same success with similar policy. IBM with Ps/2 system, Dell, HP and some other tried to force using own spare parts with minor change to otherwise standard HW and are still at it with Custom BIOS versions. I remember when Dell and HP had own custom Windows and over the shelf Windows wouldn't work.

 

[Australopithicus]

This question makes no sense. No operating system can possibly know what is going on inside the SSD at the firmware level. Garbage collection is something that the OS cannot see. All the OS can do is to send a standard ATA TRIM command to inform the SSD that particular LBA ranges are free and can be repurposed by the firmware.

Can you show us any URL where Apple claims to have special interactions with the SSD firmware?

BTW, garbage collection happens whether or not TRIM is enabled. It is a necessary consequence of wear levelling. Disabling TRIM just makes garbage collection less efficient.

My advice would be to enable TRIM and enjoy a longer life for your SSD, and better performance. Forget about this Apple/Ventura/TRIM nonsense.

The question makes perfect sense. You are jumping to conclusions. If my Apple/ Ventura/Trim concern is nonsense, why waste your time replying.

If my comments are misinformed then so is this source: Quote Hyram:

TRIM is called by your OS each time space is freed up. Garbage Collection is run by the drive’s firmware when idle and determines which parts of the drive can be cleaned up and rearranged. TRIM is obviously more efficient and obviously the drive’s firmware is aware of the TRIM commands and accounts for this.

Apple locked TRIM support for a very good reason — their code works reliably with the SSD’s they’ve chosen to use and no others, because they have programmed in nanosecond-critical timing loops that match perfectly with the access timings of the controllers used in Apple’s SSDs. Using these drivers with other controllers can, at best, slow them down, and at worst, increase the thermal effect that kills storage cells by forcing the controller to act when it isn’t quite ready. This thermal increase hastens the death of the SSD, reducing the lifespan.

 

[CountMike]

The question makes perfect sense. You are jumping to conclusions. If my Apple/ Ventura/Trim concern is nonsense, why waste your time replying.

If my comments are misinformed then so is this source: Quote Hyram:

TRIM is called by your OS each time space is freed up. Garbage Collection is run by the drive’s firmware when idle and determines which parts of the drive can be cleaned up and rearranged. TRIM is obviously more efficient and obviously the drive’s firmware is aware of the TRIM commands and accounts for this.

Apple locked TRIM support for a very good reason — their code works reliably with the SSD’s they’ve chosen to use and no others, because they have programmed in nanosecond-critical timing loops that match perfectly with the access timings of the controllers used in Apple’s SSDs. Using these drivers with other controllers can, at best, slow them down, and at worst, increase the thermal effect that kills storage cells by forcing the controller to act when it isn’t quite ready. This thermal increase hastens the death of the SSD, reducing the lifespan.

I don't think that trim by itself causes any temperature or wear increase, it's just a command that triggers other events. GC may cause some usage but it's just preemptive operation which would happen anyway when a cell marked for data deletion is attempted to be written over, so it speeds up writing.

 

[Australopithicus]

Not surprised, exclusivity with HW, OS and SW is what made Apple stand off in first place, otherwise they would be just another IBM, Dell, Hp... .. which also tried but mostly failed to have same success with similar policy. IBM with Ps/2 system, Dell, HP and some other tried to force using own spare parts with minor change to otherwise standard HW and are still at it with Custom BIOS versions. I remember when Dell and HP had own custom Windows and over the shelf Windows wouldn't work.

Yup. All about controlling the market by controlling supply lines and product exclusivity. Welcome to Capitalism.

The best Apple Mac was the G4 and the early power macs which were made accessible for upgrades. Today, in the name of security, which happens to suit marketing philosophy, Apple desktops are sealed units. That is why I like the old Intel 27" iMacs. Tricky work to remove the screen of but up-gradable just the same.

I don't think that trim by itself causes any temperature or wear increase, it's just a command that triggers other events. GC may cause some usage but it's just preemptive operation which would happen anyway when a cell marked for data deletion is attempted to be written over, so it speeds up writing.

I feel confident you are correct. Pardon me, I seem to have sent you a reply which was meant for FZABCAR. Getting my posts confused!
Thanks again for your Windows test information.
Cheers.

 

[CountMike]

Yup. All about controlling the market by controlling supply lines and product exclusivity. Welcome to Capitalism.

The best Apple Mac was the G4 and the early power macs which were made accessible for upgrades. Today, in the name of security, which happens to suit marketing philosophy, Apple desktops are sealed units. That is why I like the old Intel 27" iMacs. Tricky work to remove the screen of but up-gradable just the same.

I feel confident you are correct. Pardon me, I seem to have sent you a reply which was meant for FZABCAR. Getting my posts confused!
Thanks again for your Windows test information.
Cheers.

To top it off, SSDs are no more sensitive and expensive little things with rose petals and butterfly delicate wings but robust storage environment not needing much help. All modern OSs are aware of SSDs and their capabilities and how to use them.
I just recently switched to all M.2 SSDs (5 in all) and can't be happier for missing all those cables if nothing else.

 

[Australopithicus]

To top it off, SSDs are no more sensitive and expensive little things with rose petals and butterfly delicate wings but robust storage environment not needing much help. All modern OSs are aware of SSDs and their capabilities and how to use them.
I just recently switched to all M.2 SSDs (5 in all) and can't be happier for missing all those cables if nothing else.

I join you with that sentiment. I am very happy to see the prices coming down. But there is so much obsolete information on the web that it is not easy for an ignoramus like me to distinguish current knowledge from past. Articles are often not dated.

 

[fzabkar]

Apple locked TRIM support for a very good reason — their code works reliably with the SSD’s they’ve chosen to use and no others, because they have programmed in nanosecond-critical timing loops that match perfectly with the access timings of the controllers used in Apple’s SSDs. Using these drivers with other controllers can, at best, slow them down, and at worst, increase the thermal effect that kills storage cells by forcing the controller to act when it isn’t quite ready. This thermal increase hastens the death of the SSD, reducing the lifespan.

This is just technobabble. Please provide a reference.

 

[Australopithicus]

This is just technobabble. Please provide a reference.

 

[USAFRet]

To top it off, SSDs are no more sensitive and expensive little things with rose petals and butterfly delicate wings but robust storage environment not needing much help. All modern OSs are aware of SSDs and their capabilities and how to use them.
I just recently switched to all M.2 SSDs (5 in all) and can't be happier for missing all those cables if nothing else.

Except for the very large NAS, all my house systems have been SSD only for about a decade.

 

[Australopithicus]

OK. You may be right. It is an old post and contested post.
Source: https://web.archive.org/web/2012051...011/07/21/trim-enabler-for-lion/#comment-1957
Scroll down link to see comments. You are not the only critic.

My understanding of process now is;
1. OS operates TRIM to help SSD by confirming deleted data blocks clean and ready for Controller to write data to NAND.
2. This saves Controller Garbage Cycle doing extra work and increases SSD efficiency.
3. My impression is TRIM is triggered whenever used NAND Blocks need to be re-written.
4. Count Mike’s Crystal Disk reports TRIM on. Fzabcar advises TRIM ON helps efficiency.
5. Conclusion: TRIM ON for all SSD whether Apple or third party.

Please confirm.
Thank You.

 

[Australopithicus]

Except for the very large NAS, all my house systems have been SSD only for about a decade.

I wondered where my SUV had Gone!

 

[fzabkar]

nothing that I can see

Still no references ???

It would appear that you're a diehard Apple fanboy who believes that Apple's engineers and programmers can achieve the impossible. Perhaps a look inside some Apple engineering will enlighten you.

In the following video you will see that an "APPLE AUTHORIZED repair" for an issue due to bad soldering involves gluing a rubber sandwich to the top of the IC so that the pressure from the laptop's case overcomes the soldering fault. A real engineer would remove and reball the IC, but Apple's engineers "think different", as the company loves to remind us.

How APPLE AUTHORIZED repair refurbishes Macbooks to FAIL AGAIN! GPU kernel panics:
https://www.youtube.com/watch?v=XaGHcBZjmWA

 

[CountMike]

Expensive and exclusive doesn't always make something good. When choosing Apple it's more about OS and software/applications, whole environment that is. although some times innovative, their HW was mostly sub par when you look at raw numbers, having tight grip on whole environment enabled them to make full use of it. It's a whole other world all by itself.

 

[Australopithicus]

Still no references ???

It would appear that you're a diehard Apple fanboy who believes that Apple's engineers and programmers can achieve the impossible. Perhaps a look inside some Apple engineering will enlighten you.

In the following video you will see that an "APPLE AUTHORIZED repair" for an issue due to bad soldering involves gluing a rubber sandwich to the top of the IC so that the pressure from the laptop's case overcomes the soldering fault. A real engineer would remove and reball the IC, but Apple's engineers "think different", as the company loves to remind us.

How APPLE AUTHORIZED repair refurbishes Macbooks to FAIL AGAIN! GPU kernel panics:
https://www.youtube.com/watch?v=XaGHcBZjmWA

Still no references ???

It would appear that you're a diehard Apple fanboy who believes that Apple's engineers and programmers can achieve the impossible. Perhaps a look inside some Apple engineering will enlighten you.

In the following video you will see that an "APPLE AUTHORIZED repair" for an issue due to bad soldering involves gluing a rubber sandwich to the top of the IC so that the pressure from the laptop's case overcomes the soldering fault. A real engineer would remove and reball the IC, but Apple's engineers "think different", as the company loves to remind us.

How APPLE AUTHORIZED repair refurbishes Macbooks to FAIL AGAIN! GPU kernel panics:
https://www.youtube.com/watch?v=XaGHcBZjmWA

I replied to your post on Thursday at 9.43. With references. Scroll back and you will see it.
I take exception to your unwarrented personal remarks fzabkar. You are most offensive in your attitude. You need to grow up and learn some courtesy. Non of your presumptions are justified and your childish partisanship is entirely out of place in discussing this topic.