Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
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.
The actual physical drive install is dead simple. It's even easier than for the 2014 Mac mini. Just make sure you have a bootable copy of the Monterey installer on a USB stick in advance.

However, there are other problems to worry about. A lot of the third party SSDs use too much power, which will significantly decrease battery life. There are kernel extensions out there that can mitigate this, but they can introduce their own problems. With a Mac mini the power consumption is not a consideration since it's a desktop, always plugged in. For example, I put a Western Digital SN550 in my 2014 Mac mini with good results, but I wouldn't put that in my 2015 MacBook Pro.

Check the first page for a list of tested drives. I personally just went with another Apple OEM drive for the 2015 MacBook Pro, but they are more expensive, and they are used. The one I got had already 1/3rd of its life used, but that means it has 67% of its life still available, which should be fine for this machine since I anticipate low SSD writes on it.
The other consideration is that without a native Apple drive, future OS updates may not be possible. A recent update looked for an Apple drive to update firmware - on not finding it, the OS update would fail and unwind - at least, that is my understanding. So, keeping an Apple drive around (even if you don't use it day to day) is a good idea, if you want future updates. If the Macbook in question is 2015, then it's still getting updates, I believe.

On US ebay, 256GB SSUBX drives are going for as little as $28 (from a few days ago, I checked). If yours is an earlier model than Macbookpro11,4, then you can make do with an SSAUX drive, which is even cheaper.

To search for a SSUBX model drive, use the term "MZ-JPV256" or "MZ-JPV128" for 256G or 128G respectively. For SSUAX, search for "MZ-JPU256" or "MZ-JPU128". All the SSUBXs were made by Samsung, but the SSUAX were made by Samsung, Sandisk and Toshiba - apparently the Samsung is the one to prefer.
 
  • Like
Reactions: EugW
Hi all!

I have an MBP Retina Mid 2015 15’’ ( MacBookPro11,4) with Boot ROM Version: 428.40.10.0.0. From what I had understand I should buy an Sintech adapter If I want to buy a normal M2 NMVe SSD, it is correct?

Up to now I had the original SSD 256 and to be honest I didn’t know how fast it is, so my question is the following: since I much care about power efficiency, from the graph in the first page, Curcial P2 should be a good compromise between power efficiency and read/write performance, it is right?

I didn’t play any games, neither mount videos nor edit photo (just to say that I don’t care about P5 performance for example).

Thanks to all in advance.
I have that model. If I recall correctly, it's about 1700/1000 MB/s read/write. There are a couple of standard disk testing programs, free - download one and check. For instance:


11,4 has PCIe 3.0x4, and therefore has potential to go above 3000MB/s read/write with the right SSD.

It's not quite as simple as getting the Sintech and a random NVMe drive - there are those SSDs that work better than others and getting the right firmware update is essential for some.

If you are in the US, my strong recommendation is the SK Hynix P31, which is a very power efficient model and should allow you got get pretty good speed out of it too. But, read recent posts about the need to update the firmware on it and how you do that.
 
The other consideration is that without a native Apple drive, future OS updates may not be possible. A recent update looked for an Apple drive to update firmware - on not finding it, the OS update would fail and unwind - at least, that is my understanding. So, keeping an Apple drive around (even if you don't use it day to day) is a good idea, if you want future updates. If the Macbook in question is 2015, then it's still getting updates, I believe.
Which update was this? Anyhow, I'll keep the old 128 GB SSUBX as a backup.

On US ebay, 256GB SSUBX drives are going for as little as $28 (from a few days ago, I checked). If yours is an earlier model than Macbookpro11,4, then you can make do with an SSAUX drive, which is even cheaper.

Really? That's über cheap. I paid ~US$49 for a 256 GB SSUBX shipped from a Canadian seller, although that's because I'm in Canada. You can usually add about US$20 or so for shipping and import fees when shipping from US, which means a $28 sale becomes a ~$50 sale. Ugh.

BTW, I just got a 2017 MacBook Air too. These 2015 Pros and 2017 Airs are so cheap on the used market these days, now that the M1 Pros are out. This one also came with a 128 GB SSUBX, with only about 9% wear. As expected, it gets about 1400 MB/s read and about 700 MB/s write.

Now to install Monterey on it. It actually had a clean install of High Sierra on it. I'm assuming the last significant firmware update was for High Sierra (for NVMe support), and there were no significant firmware updates between High Sierra and Monterey, right? That way I can just directly install Monterey.
 
Hi,

It's not that your A2000 not requiring kexts. It's you _chose_ not to reduce idle power consumption further by installing kexts. These are two totally _different_ concepts. I hope people will stop false claim like this as like I pointed out to another individual poster earlier.

Pay some tribute to people before you spending many years and trying to lower power consumption with different methods. It's not the m.2 SSD vendors who take care of your use case in Mac's magically.

Yes.

I chose to not install any additional kexts, because it doesn't need any. Power consumption at idle goes down to zero and on average is always below 0.2A. It only goes up during high workload, and the temperature doesn't really go that high anymore.

So, does it need any kext? No. Because it doesn't really need it. Of course, people like you think that lowering its consumption at idle has its benefits, but I'm constantly using it, and put it to sleep or shut it down when not in use (0.0A when the display turns off on its own), so it's never really on <idle> at all.

SSDs that need the kexts are those that run hot, 0.3A on average on idle. And that's what your kext does, right? Lower current consumption on idle.

So, no need to <pay tribute> since I acknowledge the effort they've (and you've) done to help others. I've considered using the kexts before the installation if the power consumption was high, but, since it's not really that bad, I'd rather not risk doing a low-level SIP hack* just to get a few measly mA off my SSD when it's running fine without it.

Here's my original post.

*I've contributed before with the 2011 dGPU issue, hence I know the hassles and difficulties of a hack the goes into the system. It might not be an issue for you, but, like I said in my post, I'd rather not risk anything if I don't have to.

I've read a lot of posts in the thread, all mentioning high idle currents with different SSDs.

That is why I chose the A2000 to begin with, mainly because I know of Kingston's history with Apple as a third-party peripheral supplier. If I happen to get the same results of high temperature and currents like other SSDs, I would have taken the risk and used the kext.

But, like I said, the A2000 doesn't need it, at least, not in my case. Please do not take it as a slight on your effort, nor was it my intent. I'm simply stating a relative fact based on the information I have based on the results I got.

Edit.

To clarify what I've been saying, this is the SSD current in the past hour which I am actively using the computer, either browsing, typing/posting (low, 0.17A), and working with heavy graphics (peak/high 0.54A highest, 0.30A average). I'm never idle, as even if I'm not working, I'll be watching a video.

It goes to zero when the display goes to sleep in one minute. So, have a low idle current doesn't really do me much. Now, if the kext helps with overall power consumption during high use, then maybe I'll install it because then, I will need it.

Screen Shot 2021-12-17 at 13.49.03.png
 
Last edited:
  • Like
Reactions: iKalamaZoo
Hi @xanderx007

No. You still don't get it.

From your prose, I take it A2000 idle at 0.17A. I think your buddy (who I responded to earlier on a similar claim) idle aboutn 0.1xA ish as well if I recall correctly.

With 3rd party kexts, A2000 should be idling below 0.1A.

0A is bogus which I won't spend time digressing into again.
 
  • Like
Reactions: Eugene28
Hi @xanderx007

No. You still don't get it.

From your prose, I take it A2000 idle at 0.17A. I think your buddy (who I responded to earlier on a similar claim) idle aboutn 0.1xA ish as well if I recall correctly.

With 3rd party kexts, A2000 should be idling below 0.1A.

0A is bogus which I won't spend time digressing into again.

Oh I get it. You worked on it. It works for certain people who need it.
But, why are you insisting on people who can do without it? Just because you worked hard to make it?

That's just childish.

You're the one who isn't getting it. 0.07A less IS NOT a significant power saving when it's rather obvious that though 0.0A at idle might be a fluke (I mentioned it several time), the A2000 is not drawing that much power at all during idle or sleep.

It also does not have hibernation or sleep issues, or siginificant battery drain, like other SSDs, nor did I have to update the firmware since it's updated (besides the point, but I included that fact nonetheless). So I'm not worried about its power consumption, because, again, if your kext reduces overall power use, not just idle power, then by all means I'll put all effort to install, risk and all.

The A2000 worked <out of the box> without the numerous issues, hence I do I do not need the kext.

So, why are you getting worked up about something you worked on, but I do not have a use for? It's not as if I am blaming you for anything.

Take note, you have a disclaimer: You are not responsible for anything that happens using the kext.

Remember why you have that disclaimer. You are also NOT responsible if people choose not to use it.

People who have used your kext have thanked you enough. Why are you screaming at those who have not use it for recognition?

Just because I said the A2000 (at least in my case) it doesn't need it?

Thank you for your effort, but I do not need it. If I did, I'd even pay you for it, like I have with the utilities and apps I use.
 
  • Like
Reactions: iKalamaZoo
OMG. @xanderx007 you got me wrong.

I'm trying to correct people's misconception. You claim is false. Pls read my post #9902 and #9907 multiple times if you can't comprehend..

I'm getting out of here. Peace.
 
OMG. @xanderx007 you got me wrong.

I'm trying to correct people's misconception. You claim is false. Pls read my post #9902 and #9907 multiple times if you can't comprehend..

I'm getting out of here. Peace.
No. I didn't get you wrong. I've read your posts and I understand quite a lot about needing to lower power consumption. You however, cannot see it on my standpoint that lower idle power is insignificant to someone who is a heavy user, but puts the MBP to sleep or shuts it down when not in use for long periods.
 
Last edited by a moderator:
  • Like
Reactions: iKalamaZoo
Regardless how HEAVY usage a user is, most of the time your A2000 is spent idling inside your Mac laptop. It's a laptop. We aren't talking about busy server workloads here.

So A2000 consumes 0.17A at idle AS WELL AS during sleep. He he...

Bye @xanderx007
 
Regardless how HEAVY usage a user is, most of the time your A2000 is spent idling inside your Mac laptop. It's a laptop. We aren't talking about busy server workloads here.

So A2000 consumes 0.17A at idle AS WELL AS during sleep. He he...

Bye @xanderx007

No. It's not. How do I know?

I check the battery regularly since I installed it. In 5 hours of sleep, it barely moved 1%. I get 3-5 hours on medium to heavy use (Adobe apps, watching a video while working) and even more when I'm just browsing.

I also do not leave it idle for long, now, just in case you don't know what constant, HEAVY use actually means to someone who uses the laptop heavily constantly, there is barely any idle time when I work, and again, 0.07A or 70mA is barely a blip to a heavy user like me when my laptop basically is never really idle.

That is what you do not understand.
 
Last edited by a moderator:
  • Like
Reactions: iKalamaZoo
I have that model. If I recall correctly, it's about 1700/1000 MB/s read/write. There are a couple of standard disk testing programs, free - download one and check. For instance:


11,4 has PCIe 3.0x4, and therefore has potential to go above 3000MB/s read/write with the right SSD.

It's not quite as simple as getting the Sintech and a random NVMe drive - there are those SSDs that work better than others and getting the right firmware update is essential for some.

If you are in the US, my strong recommendation is the SK Hynix P31, which is a very power efficient model and should allow you got get pretty good speed out of it too. But, read recent posts about the need to update the firmware on it and how you do that.
Thanks a lot for your reply.

I’m from Italy and to be honest I didn’t know if I can buy the SK Hynix P31 from anywhere.
Previously I was oriented towards the P2 but reading some post here in this thread I figure out that there will be kernel panic and other problems with it. Now I’m doing some research on the Sabrent Rocket or SX8200 Pro.

What do you suggest?
 
Just to clarify..

I probably would be around this thread for a long while. And will try to engage in conversations that I pick or when I have time in the usual manner like how I conducted in the past.

So people expect me to disappear will be disappointed. New people expect me to engage in long messages perhaps will also be disappointed. Personally I can't expect one to explain the same stuff again and again. There are good educators out there for nothing or a living. Or there is google to mine the wealth in this thread.

Peace.
 
Regardless how HEAVY usage a user is, most of the time your A2000 is spent idling inside your Mac laptop. It's a laptop. We aren't talking about busy server workloads here.

So A2000 consumes 0.17A at idle AS WELL AS during sleep. He he...

Bye @xanderx007
No. It goes down to ZERO during idle and sleep.
 
Last edited by a moderator:
  • Like
Reactions: iKalamaZoo
Isn't this the reason why in your updated list, the A2000 installed in an 11.4 is no longer included because the power saving is insignificant?



I learned 70mA from you in one of your posts today. So either you read about someone tried it with 3rd party kexts or yourself had tried it. Before today I wasn't aware of this 70mA figure.

I used to check this thread quite frequently when ppl reported results with ssdpmEnabler. I made a judgement on user's test method and usually will include it unless there are many duplicated entries in Confirmed Working Models list already.

I was not aware of anyone reporting A2000 result with ssdpmEnabler. That's why it's not on the list...(unlike how you thought which shows me you're childish again..)

As a matter of fact, the first time I learned A2000 was from your buddy..

Anyway, as I said I care less about what you use. Your strong assertion of A2000 not needing kext is simply false. Pls stop misleading others.
 
Last edited by a moderator:
  • Like
Reactions: Eugene28
Hi,

It's not that your A2000 not requiring kexts. It's you _chose_ not to reduce idle power consumption further by installing kexts. These are two totally _different_ concepts. I hope people will stop false claim like this as like I pointed out to another individual poster earlier.

Pay some tribute to people before you spending many years and trying to lower power consumption with different methods. It's not the m.2 SSD vendors who take care of your use case in Mac's magically.

While I think it is in general a good idea to keep your posts short and easy to read for everybody, I also think it is quite easy to read between the lines and take people’s speculations with a grain of salt.

One inadvertently misleading claim that I have read here sometimes was that you do or don’t need a kext like ssdPMenabler. First, let’s make it clear that any SSD on the list on page 1 and on kvic’s compatibility list will work without any special kext, if the Mac’s Bootrom and OS is new enough. The compatibility started at some point with High Sierra (still works fine on my Macbook Pro) and has somewhat improved with Mojave. So it is safe to say that if you have installed Mojave or higher, you are good to go.

It is very understandable that people want to buy an SSD that has a low power consumption out of the box, but Apple has tuned their SSDs so they are consuming less power than NVMe SSDs. So it would be indeed interesting to find out how much power can be saved by using ssdPMenabler. Installing a kext like ssdPMenabler or NVMEfix will reduce power consumption for sure. If you want the best battery runtime for your Macbook, simply try it for yourself.
The website with it’s install and uninstall guide is quite comprehensive.

Let’s get back to the numbers, and how to interpret them. Is the Kingston A2000 really consuming less power than other low budget DRAM-less SSDs? I am curious to find out.

Beginning with the 0,17 A mentioned here, I see the number is very close to the 0,18 A (600mW) measured by Tom’s Hardware on the bottom of that page:
This is exactly the number that the table on page 1 shows:

So, assuming that ssdPMenabler makes use of Active-state power management (ASPM), the A2000 would really consume a lot less power when using that kext. Ideally that would be 0.0036 A (12 mW) according to Tom’s Hardware.

In comparison, Tom’s Hardware measured about 0.24 A (800 mW) for the WD SN500 or 0.13 A (434 mW) for the Corsair MP510. As for the WD the numbers are matching well with the table on page 1, while the Corsair uses a little more compared to the Tom’s article.
I get the impression that the table on page 1 is quite accurate and can well be used to evaluate SSDs power consumption in comparison.

As you can read from kvic’s confirmed working Mac models page, the Corsair can save a lot power and also can the WD.

So far I have learned from these numbers that out of the box the A2000 is not bad, but also not much better than others. Of course the power consumption during workload (*with photo editing, watching videos and web surfing the SSD consumes less power than with bench tests) is another factor. But then in a Macbook Pro 11,4 or 11,5 the Corsair would be faster.

This is still much theory, but the only way to find out is running tests and comparing real numbers from a Macbook Pro. Maybe the A2000 will improve much with ssdPMenabler, who knows?
I would find it very interesting to see more Macbook models and SSDs appear in kvic’s compatibility list.

Btw, posts can be edited here any time, in case you guys want to trim your very long discussion a bit ;-)
 
Last edited:
Forgive me if this has been discussed before already, but I feel the need to mention again that the data in the table on the first page for the Apple OEM drives may not be correct. Or at least, it is misleading. It states 0.4 Watts at idle for the 512 GB UBX SSD (MZ-JPV5120/A04), but I get 10 mA / 0.03 Watts at idle for the 256 GB UBX SSD (MZ-JPV2560/A04). Also, my read consumption was about 1.5 Watts (although I'd have to test that further to be sure).

However, that table was testing the drives with the 2013 rMBP 13". This may not be an appropriate test, since that machine never shipped with these drives AFAIK. I am using a 2015 rMBP 13" which shipped with these drives. Perhaps Apple never bothered to implement the same power management features with mismatched drives vs MacBook Pros. It does make me wonder how much power the older UAX drives draw in those older MacBook Pros. Maybe if you matched a slower Apple OEM drive to a pre-2015 machine that actually shipped with that slower drive, it'd use the right power management features already built into macOS?

I also note that the write speed for that 512 GB drive is listed in the table as 700 MB/s write and 1150 MB/s read for that drive in a 2013 MBP, which negatively skews the efficiency rating even further. With my 256 GB model, I get 1220 MB/s write and 1390 MB/s read... but in a 2015 MacBook Pro. It's only my 128 GB model (MZ-JPV1280/A04) that drops to about 700 MB/s write.
 
Last edited:
If you are in the US, my strong recommendation is the SK Hynix P31, which is a very power efficient model and should allow you got get pretty good speed out of it too. But, read recent posts about the need to update the firmware on it and how you do that.
It looks like the SK Hynix P31 is also available in the UK at a good price with 500GB for £77.32 & 1TB for £134.35.

 
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
I tried a Sabrent Rocket 2TB in my Mid 2015 15" MBP but discovered that write performance dropped off drastically once the write buffers were full. Initially performance looked great but after five cycle of Black Magic Disk Speed Test write performance dropped to less than half the initial value.

I replaced the Sabrent Rocket with a 2TB Apple Polaris that I bought new for about $800. Initial performance is down on the Sabrent Rocket but there is no drop in write performance even if I let it run for five minutes. I'm very happy with the performance & battery life is much better (5 hours or more versus a maximum of 3 hours).
 

Attachments

  • Screenshot 2021-12-18 at 11.27.48.png
    Screenshot 2021-12-18 at 11.27.48.png
    988.1 KB · Views: 72
Zkoušel jsem Sabrent Rocket 2TB ve svém 15" MBP z poloviny roku 2015, ale zjistil jsem, že výkon zápisu drasticky poklesl, jakmile byly vyrovnávací paměti pro zápis plné. Zpočátku výkon vypadal skvěle, ale po pěti cyklech testu Black Magic Disk Speed Test klesl výkon zápisu na méně než polovinu. počáteční hodnota.

Vyměnil jsem Sabrent Rocket za 2TB Apple Polaris, který jsem koupil nový za asi 800 $. Počáteční výkon je u Sabrent Rocket nižší, ale nedochází k žádnému poklesu výkonu zápisu, i když jej nechám běžet pět minut. S výkonem jsem velmi spokojen a výdrž baterie je mnohem lepší (5 hodin nebo více oproti maximálně 3 hodinám).
Where did you buy a new Apple SSD?
 
It looks like the SK Hynix P31 is also available in the UK at a good price with 500GB for £77.32 & 1TB for £134.35.

They don’t sent in Italy.

Other recommendation for others SSDs? Up to now I just bought the adapter.
 
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
Which update was this? Anyhow, I'll keep the old 128 GB SSUBX as a backup.



Really? That's über cheap. I paid ~US$49 for a 256 GB SSUBX shipped from a Canadian seller, although that's because I'm in Canada. You can usually add about US$20 or so for shipping and import fees when shipping from US, which means a $28 sale becomes a ~$50 sale. Ugh.

BTW, I just got a 2017 MacBook Air too. These 2015 Pros and 2017 Airs are so cheap on the used market these days, now that the M1 Pros are out. This one also came with a 128 GB SSUBX, with only about 9% wear. As expected, it gets about 1400 MB/s read and about 700 MB/s write.

Now to install Monterey on it. It actually had a clean install of High Sierra on it. I'm assuming the last significant firmware update was for High Sierra (for NVMe support), and there were no significant firmware updates between High Sierra and Monterey, right? That way I can just directly install Monterey.
You'll get a firmware update with Monterey.. Actually, I think there have been several since High Sierra.
 
  • Like
Reactions: EugW
You'll get a firmware update with Monterey.. Actually, I think there have been several since High Sierra.
So are you saying it's safe to go straight from High Sierra to Monterey, because Monterey will have the latest update?
 
Where did you buy a new Apple SSD?
Inevitably it was on eBay. I bought my 2TB Polaris SSD over a year ago for my 2014 MBP because I filled up the 1TB SSUBX I had fitted previously. I just bought a Mid 2015 MBP for a great price & have moved the 2TB Polaris over. With PCI 3.0 in the 2015 MBP the performance is considerably higher than in the 2014 MBP. This is where I got my 2TB Polaris:- https://www.ebay.co.uk/itm/352638228107
 
Dear All,I have been studying this topic for so long but cudnt find answers to my problem.I am certainly missing some points here to address my issue.I have a Macbook Pro Early 2013 A1398 which has 7+17 pins on the ssd slot.I purchased a Seagate FireCuda 520 1TB Performance Internal Solid State Drive SSD PCIe Gen4 x4 NVMe 1.3 thinking a riser/adapter would solve my issue.However its has been a never ending search to find an adapter for it.If anyone can help me with a link for the adapter would be great.If this NVMe SSD is not compatible with my macbook early 2013 kindly help which all ssd's can be used directly without an adapter or with an adapter.
There are 7+17 adapters, but they are either for m.2 sata or for msata.

For instance:




I assume that the first or third adapter would allow you to use an m.2 SATA drive such as:


The second would let you use an msata drive - those don't seem to be manufactured by well-known brands anymore.
 
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.