Someone has here. I don't foresee a problem with the Sintech adapter as there is plenty of clearance between the Sintech adapter and my 480GB, which has chips on both sides and between both and the MBP. Possibly with a MBA as space is tighter.
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.
Upgrading 2013-2015 Macbook Pro SSD to M.2 NVMe
- Thread starter maxthackray
- WikiPost WikiPost
- Start date
- Sort by reaction score
- Status
Someone has here. I don't foresee a problem with the Sintech adapter as there is plenty of clearance between the Sintech adapter and my 480GB, which has chips on both sides and between both and the MBP. Possibly with a MBA as space is tighter.
Oddly - I've now started to get kernel panics. I suspect it may be the nvme adapter (I've got a CC - waiting on the Sintech to be delivered) but its happening when the MacBook is going to sleep. The only change I remember was changing hibernate mode to 25; so I've reset via power settings and will wait and see if it acts up again.
My MacBook Air 2014 with black sintech adapter and 512gb 970 evo it fits but does touch the bottom cover. I think the air has less clearance. The plug that the 970 evo plugs into is the highest point.
I’m not 100% sure the bottom cover goes on completely as the center clip on the battery is broke on mine but all the edges and screws are seated.
Great to know! Just to make sure, do you have the *long* black sintech adaptor?
That's pretty amazing. I did the same test with the WD Black G2, in fact with hibernatemode set to 25. My MBP lost 9% charge in 11 hours.
But didn't you report 6% drain for a similar test in your first post with the SX8200? Just curious about the difference.
Also surprised to hear no sleep issue in pre-2015 MBP. The differences seem to be that you have Mojave PB and a JSER adapter. Maybe one of these things (or their combination) fixes the sleep issues with aftermarket SSDs!
Last edited:
Yup: weird, huh? I honestly have no explanation. I will keep testing and see how things go. Also I just trust iStatMenu. Is that reasonable? I do not know but think so.
Anyway, regarding battery drain: yesterday at 19:49: 100% charge. I closed the lid. Today at 8:11 opened the lid, woke up with no issue, 99% charge, closed the lid. Today at 18:10 opened the lid, woke up with no issue, 95% charge.
This would show the WD Black is not-so-great as most people say it is, while the intel 760p is far better. right ?
That's just great. I know you mentioned high temps and power draw with the Samsung, but did you also check the sleep battery drain with it? If Samsung had high drain (3% per hour), and everything else stayed the same except SX8200, it would indicate that changing the SSD is the only factor that improves everything.
Anyway, I already ordered an SX8200 after seeing your story and it will arrive tomorrow. I'll report what I see as well.
Also very happy for you that it currently seems like smooth sailing for you. It would be fantastic if Mojave PB/GM ends up addressing our issues as it shouldn’t be too long before it’s officially released.
Did you perform the command sudo pmset -a standby 0 ? I saw in your first first it actually said standby 1?
Something is changed in Mojave. There are new parameters such as standbydelaylow/high, Standby Battery Threshold. I think these can make your macbook sleep and wake safely without KP. Of course Apple might do something in Mojave for using NVMe ssd in 2013-2014 macbook.
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
Yeah: power draw of the Samsung at idle was 0.47A and that showed in battery life. I was getting around 4 hours. Also remember that on a fresh install macOS does a lot of caching so battery life at first will be crap. After a few hours everything goes back to normal.
[doublepost=1536040945][/doublepost]
Nope. I left everything to default values
The SX8200 arrived. Testing it right now. I left the pmset hibernation in 25 for the time being. I can tentatively confirm that the power usage idling is circa 0.19. Here is a screenshot. I will report tomorrow about how long the battery lasts.
Attachments
Hi @ all,
I am reading this Thread since month now.
I also have a mid 2014 MacBook Pro. I just installed a 970 EVO 1 TB. Macbook Pro started without any issues in the recovery mode. The disc was recognized.
Now the MacBook is restoring from a time machine back up.
I did not change any settings regarding sleep and so on. I will test the disc and battery life with the actual settings so that i can compare the disc.
Will keep you update.
Many Thanks to all of you with great support, ideas and help.
Andy
I am reading this Thread since month now.
I also have a mid 2014 MacBook Pro. I just installed a 970 EVO 1 TB. Macbook Pro started without any issues in the recovery mode. The disc was recognized.
Now the MacBook is restoring from a time machine back up.
I did not change any settings regarding sleep and so on. I will test the disc and battery life with the actual settings so that i can compare the disc.
Will keep you update.
Many Thanks to all of you with great support, ideas and help.
Andy
I have a similar setup, rMBP 15" Mid-2014 with Samsung 970 but I have 500Gb SSD. I have no issues with the setup at all, I did a clean installation.
However, I am facing the sleep/hibernation problem, sometimes happens and sometimes doesn't. Does anyone know something to try to fix it? Reset SMC or things like this work?
I am using the SSD with the Sintech Black adapter (ST-NGFF2013-C)
Cheers
Ok, after a day of testing , battery life went down from 5 hours with the apple stock ssd to 4 hours with the sx8200 (250gb) , there is also more heat (I understand this should not be an issue) probably because all the Mac is aluminium. All in all , it is a trade of 1 hour battery life for more storage at a reasonable price. I think I will keep it as this.
So the SX8200 came in yesterday. The installation etc. worked normally, except the SX8200 doesn't format with 4K sectors, unlike the WD. I'm not sure if this is going to be a problem in the future.
As reported here, the idle current draw is 0.19A, about half of the WD Black G2. On rMBP 13" Early 2015, the read speed is 1500 MB/s and write speed is about 1100 MB/s. I must note that sustained write is actually much faster on the SX8200 as compared to WD G2, which starts out at 1200-1300 MB/s and dies down to 600 MB/s within 10 seconds. The temperatures are also lower with the SX8200.
The battery issue is still open though. In one three hour sleep, the MBP drained 16% (!) of battery. An overnight sleep of 7 hours only resulted in 4% battery drain. After some digging, I think the cause of battery drain is because Mac keeps waking up every 10-15 minutes to perform some network operation. This is a reported issue on line, and could be fixed with setting tcpkeepalive to 0. For some reason, Mac stopped waking up after 1 hour of sleep when left overnight, and thus the low battery drain.
As a followup, I would appreciate if @ilovejamon could report what his tcpkeepalive setting is. Some output from pmset -g log would be even better to see if his MBP is waking up frequently too.
Hi, please see my post #1909 for the output of pmset -g. The value of tcpkeepalive does not appear which I guess means that it is set to its default value (which is 1 if I am not mistaken). I should say that I am using a different adapter (JSER) than most people here and that I put kapton tape on the adapter pins. I do not know if these things have any effect on battery drain. Also I have PowerNap disabled while on battery (which I believe is the default).
Lastly, I am on Mojave public beta.
Yes, thanks. I had noticed that tcpkeepalive was missing from pmset -g, hence the request.
I wonder if Apple has set default to 0 in Mojave since many people complained? Anyway, I would be interested in seeing the output ofpmset -g log | grep -e " Sleep " -e " Wake " -e " DarkWake "
after an overnight sleep.
Not sure. In one of his posts, @gilles_polysoft remarked that Samsung SSDs kernel panic when writing very large files, and he conjectured that it might be because Samsung SSDs can't be formatted to 4K sectors (stock Apple SSDs have 4K sectors).
Okay, I might have failed to notice that it got solved with APFS. So 512 byte sectors are going to be fine, great!
Yes, thanks. I had noticed that tcpkeepalive was missing from pmset -g, hence the request.
pmset -g log | grep -e " Sleep " -e " Wake " -e " DarkWake "
after an overnight sleep.
Sure thing. I have been keeping the laptop plugged in the whole time the last few days. However it was on battery during the night between Sep. 2 and Sep. 3. Here is the output you requested:
2018-09-02 19:52:55 +0200 Sleep Entering Sleep state due to 'Clamshell Sleep': Using Batt (Charge:100%) 44455 secs
2018-09-02 19:53:12 +0200 PM Client Acks Delays to Sleep notifications: [com.apple.apsd is slow(17437 ms)]
2018-09-03 08:13:50 +0200 Kernel Client Acks Delays to Sleep notifications: [AppleIntelFramebuffer driver is slow(msg: SetState to 1)(408 ms)] [powerd is slow(17440 ms)] [AppleThunderboltNHIType2 driver is slow(msg: WillChangeState to 2)(366 ms)] [RP05 driver is slow(msg: SetState to 0)(902 ms)]
2018-09-03 08:13:50 +0200 Wake Wake from Normal Sleep [CDNVA] due to EC.LidOpen/Lid Open: Using BATT (Charge:99%) 31 secs
Great, thanks. As you can see, there are no wake ups between 19:53:12 and 08:13:50. By contrast, when connected to WiFi, my rMBP wakes up every 10-15 minutes with lid closed. This is the sole cause of all battery drain for me because if I turn WiFi off the rMBP drains no battery in the last 2.5 hours, and I think also explains why you got such great results.
The reason for frequent wake ups is the only mystery now. I will not change the default settings for a couple of days and observe the behavior, to rule out any effects due to a fresh OS install. If it still doesn't sleep well after a few days, I might experiment with tcpkeepalive.