It would be interesting to note if there is much of a real world difference between the 2.2 and 2.6 model one since there really is only a $100 difference between similarly optioned models. ($300 between 2.6 and i9).
We'll be talking about this a bit tomorrow.
Here's a pair of questions to try to clarify the power-vs-performance balance: at idle (email/writing text/web browsing) is the 2.2Ghz + 555X noticeably more thrifty with power than others? And at full chat (continuous compiling/transcoding/whatever), what is the _actual_ sustained speed for the 2.2 vs. 2.6, i9? Those are two questions which no review has touched.
[Perhaps we would not gain much (~10%?) in "compile farm" performance with the faster CPUs because the sustained all-core speed is capped by the thermal envelope? On the other hand we might not gain much battery life because all CPUs throttle back very well these days. ]
Vadim (one of AI's video guys) has addressed this a bit in the post right above yours.
Battery life measurements over a week beyond generalizations are useless and nowhere near enough time to get a good set of data regarding percentages and whatnot. We'll be revisiting most of these machines in about six months regarding that.
Vadim's measurements were probably enough for me to narrow down my choice, thanks! A few percent performance gain, even with the larger cache? Not so tempting.
I used the term "battery life", but meant to ask how long the thing runs on battery with minimal load. Have you thought about trying to figure out how much extra power the non-LP RAM consumes? If only to guess at the decisions Apple might have been making putting the 32G option off.
Huge bezels in 2018 look ridiculous and lack of FaceID is a HUGE MISS! Nobody wants yesterday's technology in a premium priced MBP! Why do back to TouchID after you already have a iPhone X?!
Also odd to me is the lack of a cellular option or slot, why offer it in iPad but not MBPs?
Do these support MU-MIMO?
I agree! Rarely do I hear anyone comment on the lack of FaceID in the new MacBook Pro. To me, this is a BIG MISS. Another sign that Apple isn't focusing on the Macs. And where is the cellular option? Should that be so hard to do?
And don't get me started on the keyboard. Apple knows better and could do better with the keyboard. It almost seems like intentional self-sabotage, given that Apple can't outright kill the MacBook Pro line. Instead, take a key aspect of the laptop -- the keyboard -- and make it uncomfortable.
Quick question to everyone: I got the 2.6Ghz 15" (2018), with 16Gb ram and 512Gb SSD; my write speeds never go past 1750MB/s (running Blackmagic Disk Speed Test), is that normal? Anyone else with the same specs can get the advertised 2500MB/s speeds? Should I be worried?
I feel like the 512Gb SSD's is slower (no RAID 0?) and nowhere is this being mentioned (not on Apple's website, not on any reviews I've come across). I bought this config thinking it'd be the 'fastest SSD ever on a notebook' (as some have touted), but I'm feeling disappointed and a bit cheated by the marketing/hype. In the end my SSD on a high end machine is apparently slower than on a 13" Macbook (all tests made on the 512Gb variant of the 13" mention high write speeds) and as slow as last year's...
I'd love to get some clarification on that and I think it would make sense if Appleinsider helped clarify and spread the word on this to other potential buyers (to avoid they fall into the trap I seem to have gone into). I certainly read a lot about this machine here before making my purchase and I would have loved to know this in advance.
Quick question to everyone: I got the 2.6Ghz 15" (2018), with 16Gb ram and 512Gb SSD; my write speeds never go past 1750MB/s (running Blackmagic Disk Speed Test), is that normal? Anyone else with the same specs can get the advertised 2500MB/s speeds? Should I be worried?
I feel like the 512Gb SSD's is slower (no RAID 0?) and nowhere is this being mentioned (not on Apple's website, not on any reviews I've come across). I bought this config thinking it'd be the 'fastest SSD ever on a notebook' (as some have touted), but I'm feeling disappointed and a bit cheated by the marketing/hype. In the end my SSD on a high end machine is apparently slower than on a 13" Macbook (all tests made on the 512Gb variant of the 13" mention high write speeds) and as slow as last year's...
I'd love to get some clarification on that and I think it would make sense if Appleinsider helped clarify and spread the word on this to other potential buyers (to avoid they fall into the trap I seem to have gone into). I certainly read a lot about this machine here before making my purchase and I would have loved to know this in advance.
Thanks
The Blackmagic Disk Speed Test is designed to test/stress the SSD device the same way video does. It does not seem to really indicate exactly what tests it is running (compressed data vs non-compressed, sequential read/write vs random access reading and writing, block size of the data etc). SSD specifications (i.e. package specs) only tend to publish the specs that you will get the highest rating from... which I believe is probably sequential read and then sequential write-only.
Quick question to everyone: I got the 2.6Ghz 15" (2018), with 16Gb ram and 512Gb SSD; my write speeds never go past 1750MB/s (running Blackmagic Disk Speed Test), is that normal? Anyone else with the same specs can get the advertised 2500MB/s speeds? Should I be worried?
I feel like the 512Gb SSD's is slower (no RAID 0?) and nowhere is this being mentioned (not on Apple's website, not on any reviews I've come across). I bought this config thinking it'd be the 'fastest SSD ever on a notebook' (as some have touted), but I'm feeling disappointed and a bit cheated by the marketing/hype. In the end my SSD on a high end machine is apparently slower than on a 13" Macbook (all tests made on the 512Gb variant of the 13" mention high write speeds) and as slow as last year's...
I'd love to get some clarification on that and I think it would make sense if Appleinsider helped clarify and spread the word on this to other potential buyers (to avoid they fall into the trap I seem to have gone into). I certainly read a lot about this machine here before making my purchase and I would have loved to know this in advance.
Thanks
The Blackmagic Disk Speed Test is designed to test/stress the SSD device the same way video does. It does not seem to really indicate exactly what tests it is running (compressed data vs non-compressed, sequential read/write vs random access reading and writing, block size of the data etc). SSD specifications (i.e. package specs) only tend to publish the specs that you will get the highest rating from... which I believe is probably sequential read and then sequential write-only.
Regardless of what the type of test Blackmagic does, shouldn’t it be consistent for all 2018 MBPs with the same capacity/number of NAND chips from the same vendor?
Quick question to everyone: I got the 2.6Ghz 15" (2018), with 16Gb ram and 512Gb SSD; my write speeds never go past 1750MB/s (running Blackmagic Disk Speed Test), is that normal? Anyone else with the same specs can get the advertised 2500MB/s speeds? Should I be worried?
I feel like the 512Gb SSD's is slower (no RAID 0?) and nowhere is this being mentioned (not on Apple's website, not on any reviews I've come across). I bought this config thinking it'd be the 'fastest SSD ever on a notebook' (as some have touted), but I'm feeling disappointed and a bit cheated by the marketing/hype. In the end my SSD on a high end machine is apparently slower than on a 13" Macbook (all tests made on the 512Gb variant of the 13" mention high write speeds) and as slow as last year's...
I'd love to get some clarification on that and I think it would make sense if Appleinsider helped clarify and spread the word on this to other potential buyers (to avoid they fall into the trap I seem to have gone into). I certainly read a lot about this machine here before making my purchase and I would have loved to know this in advance.
Thanks
The Blackmagic Disk Speed Test is designed to test/stress the SSD device the same way video does. It does not seem to really indicate exactly what tests it is running (compressed data vs non-compressed, sequential read/write vs random access reading and writing, block size of the data etc). SSD specifications (i.e. package specs) only tend to publish the specs that you will get the highest rating from... which I believe is probably sequential read and then sequential write-only.
Regardless of what the type of test Blackmagic does, shouldn’t it be consistent for all 2018 MBPs with the same capacity/number of NAND chips from the same vendor?
Probably, though the question would be is there a difference depending on capacity.
It would be interesting to note if there is much of a real world difference between the 2.2 and 2.6 model one since there really is only a $100 difference between similarly optioned models. ($300 between 2.6 and i9).
From our testing: 4 run average of Cinebench R15 CPU test back-to-back:
Base: 991 Mid: 1001 i9: 1011
Clock speed averages for each under load after heating up and stabilizing: Base: 3.05Ghz Mid: 3.1Ghz i9: 3.15Ghz
Can you clarify which systems (CPUs) are base, mid, and i9 here?
The base model with the 6-core 2.2 GHz CPU averaged 3.05 GHz? If so, Intel put this chip into the wrong bin? These Cinebench scores are all of +/- 1% different. Why even bother with higher end models if so?
Then, how often does the i9 actually hit 4.8 GHz on one core? For all practical purposes, never? The bigger cache on the i9 looks like it isn’t even helping for this benchmark.
Each chip is hugely limited by the cooling system. It's extremely hard to fit a good cooling system in that thin of a frame, with those slim vents.That's why there's such a small difference between each chip. I believe the highest Boost clock we saw was 4.2GHz for the i9 (all 6 running, of course) The issue is that people are fooled by that incredibly high Boost clock speed. The conditions have to be extremely specific for one core to reach that high.
We'll have more information on this when our 15" i9 review goes live tomorrow. Make sure to check it out.
Apart from truetone and a sleeve, this really IS just a spec upgrade - as per your review. Same keyboard shortcomings improvisingly repaired, same lack of MagSafe, same functionality - struggling to keep up with the competition RAM & graphics wise. OK, the i9 is a beast but it has to be seen if the less than average cooling support system can handle it
Apart from truetone and a sleeve, this really IS just a spec upgrade - as per your review. Same keyboard shortcomings improvisingly repaired, same lack of MagSafe, same functionality - struggling to keep up with the competition RAM & graphics wise. OK, the i9 is a beast but it has to be seen if the less than average cooling support system can handle it
Yes, without the 3D holographic imaging system built into each laptop and the holographic storage... we are left with just spec bumps
This just in: my friend who got the 15" i9 variant with 32Gb of RAM and 512Gb SSD is also seeing ~1700Mb/s.
(Maybe it's just me, but this seems like a hugely important info)
The Blackmagic Disk Speed Test is designed to test/stress the SSD device
the same way video does. It does not seem to really indicate exactly
what tests it is running (compressed data vs non-compressed, sequential
read/write vs random access reading and writing, block size of the data
etc). SSD specifications (i.e. package specs) only tend to publish the
specs that you will get the highest rating from... which I believe is
probably sequential read and then sequential write-only.
I used Blackmagic because that's what I saw on most (if not all) of the reviews out there. If anyone has any other (free/open) suggestions, or other ways of testing the SSD speed, I'd be glad to try them.
Regardless of what the type of test Blackmagic does, shouldn’t it be
consistent for all 2018 MBPs with the same capacity/number of NAND chips
from the same vendor?
Exactly my thoughts. What's weirder is that the 512Gb 13" seems to reach the advertised 2500Mb/s (at least that's what I gather from the reviews).
Is there a way of knowing for sure the number of NAND chips on my machine?
Probably, though the question would be is there a difference depending on capacity.
The answer seems to be yes, unfortunately. It appears that for some (bizarre) reason the 13" gets two 256Gb in RAID0 and the 15" gets only one with half the speed (which would be total nonsense considering not only that the 13" is cheaper, but also that the 15" would be 'more pro').
I wouldn't mind it being slower as long as it were clearly stated on Apple Store's configuration page (I even looked for it before making my purchase! The only thing mentioned is the difference in capacity, never speed).
All in all, after holding back my purchase by the whole throttling fiasco, this is very frustrating. If I had been informed in advance I'd have gotten the 1Tb model, it's not only more storage it's twice as fast (apparently).
Apart from truetone and a sleeve, this really IS just a spec upgrade - as per your review. Same keyboard shortcomings improvisingly repaired, same lack of MagSafe, same functionality - struggling to keep up with the competition RAM & graphics wise. OK, the i9 is a beast but it has to be seen if the less than average cooling support system can handle it
Sure. Apart from quad-core processors in the machine for the first time, True Tone, a new Thunderbolt 3 controller, an improved keyboard, new speakers, more RAM.
This just in: my friend who got the 15" i9 variant with 32Gb of RAM and 512Gb SSD is also seeing ~1700Mb/s.
(Maybe it's just me, but this seems like a hugely important info)
The Blackmagic Disk Speed Test is designed to test/stress the SSD device
the same way video does. It does not seem to really indicate exactly
what tests it is running (compressed data vs non-compressed, sequential
read/write vs random access reading and writing, block size of the data
etc). SSD specifications (i.e. package specs) only tend to publish the
specs that you will get the highest rating from... which I believe is
probably sequential read and then sequential write-only.
I used Blackmagic because that's what I saw on most (if not all) of the reviews out there. If anyone has any other (free/open) suggestions, or other ways of testing the SSD speed, I'd be glad to try them.
Regardless of what the type of test Blackmagic does, shouldn’t it be
consistent for all 2018 MBPs with the same capacity/number of NAND chips
from the same vendor?
Exactly my thoughts. What's weirder is that the 512Gb 13" seems to reach the advertised 2500Mb/s (at least that's what I gather from the reviews).
Is there a way of knowing for sure the number of NAND chips on my machine?
Probably, though the question would be is there a difference depending on capacity.
The answer seems to be yes, unfortunately. It appears that for some (bizarre) reason the 13" gets two 256Gb in RAID0 and the 15" gets only one with half the speed (which would be total nonsense considering not only that the 13" is cheaper, but also that the 15" would be 'more pro').
I wouldn't mind it being slower as long as it were clearly stated on Apple Store's configuration page (I even looked for it before making my purchase! The only thing mentioned is the difference in capacity, never speed).
All in all, after holding back my purchase by the whole throttling fiasco, this is very frustrating. If I had been informed in advance I'd have gotten the 1Tb model, it's not only more storage it's twice as fast (apparently).
The 512GB i9 version we're publishing the review for shortly exceeds 2500 MB/sec.
How long have you had the machine? There is a pile of indexing the drive after a new user migrates data that can run for days, and this will impact SSD benchmarks.
This just in: my friend who got the 15" i9 variant with 32Gb of RAM and 512Gb SSD is also seeing ~1700Mb/s.
(Maybe it's just me, but this seems like a hugely important info)
The Blackmagic Disk Speed Test is designed to test/stress the SSD device
the same way video does. It does not seem to really indicate exactly
what tests it is running (compressed data vs non-compressed, sequential
read/write vs random access reading and writing, block size of the data
etc). SSD specifications (i.e. package specs) only tend to publish the
specs that you will get the highest rating from... which I believe is
probably sequential read and then sequential write-only.
I used Blackmagic because that's what I saw on most (if not all) of the reviews out there. If anyone has any other (free/open) suggestions, or other ways of testing the SSD speed, I'd be glad to try them.
Regardless of what the type of test Blackmagic does, shouldn’t it be
consistent for all 2018 MBPs with the same capacity/number of NAND chips
from the same vendor?
Exactly my thoughts. What's weirder is that the 512Gb 13" seems to reach the advertised 2500Mb/s (at least that's what I gather from the reviews).
Is there a way of knowing for sure the number of NAND chips on my machine?
Probably, though the question would be is there a difference depending on capacity.
The answer seems to be yes, unfortunately. It appears that for some (bizarre) reason the 13" gets two 256Gb in RAID0 and the 15" gets only one with half the speed (which would be total nonsense considering not only that the 13" is cheaper, but also that the 15" would be 'more pro').
I wouldn't mind it being slower as long as it were clearly stated on Apple Store's configuration page (I even looked for it before making my purchase! The only thing mentioned is the difference in capacity, never speed).
All in all, after holding back my purchase by the whole throttling fiasco, this is very frustrating. If I had been informed in advance I'd have gotten the 1Tb model, it's not only more storage it's twice as fast (apparently).
The 512GB i9 version we're publishing the review for shortly exceeds 2500 MB/sec.
How long have you had the machine? There is a pile of indexing the drive after a new user migrates data that can run for days, and this will impact SSD benchmarks.
I got it five days ago and I didn't migrate any data, plus I don't see mds or any suspicious processes running while testing the SSD. I even tried turning off FileVault (and saw no difference). I've even tried the test on different days before posting here.
Huge bezels in 2018 look ridiculous and lack of FaceID is a HUGE MISS! Nobody wants yesterday's technology in a premium priced MBP! Why do back to TouchID after you already have a iPhone X?!
Also odd to me is the lack of a cellular option or slot, why offer it in iPad but not MBPs?
Do these support MU-MIMO?
Not everyone has or indeed wants an iPhone X. TouchID is perfect for me and I really don't like FaceID. I'll be one of those NOT buying a phone with FaceID on it for as long a possible. I have specific reasons for not wanting to use FaceID but would like a quick method of logging in.
I would immediately buy this model if offered with function keys (and a $200 discount). Now I am more reluctant to do that. Almost all programs I need to use frequently have no or very limited support for the touchbar, while I really need a decent ESC key
This just in: my friend who got the 15" i9 variant with 32Gb of RAM and 512Gb SSD is also seeing ~1700Mb/s.
(Maybe it's just me, but this seems like a hugely important info)
The Blackmagic Disk Speed Test is designed to test/stress the SSD device
the same way video does. It does not seem to really indicate exactly
what tests it is running (compressed data vs non-compressed, sequential
read/write vs random access reading and writing, block size of the data
etc). SSD specifications (i.e. package specs) only tend to publish the
specs that you will get the highest rating from... which I believe is
probably sequential read and then sequential write-only.
I used Blackmagic because that's what I saw on most (if not all) of the reviews out there. If anyone has any other (free/open) suggestions, or other ways of testing the SSD speed, I'd be glad to try them.
Regardless of what the type of test Blackmagic does, shouldn’t it be
consistent for all 2018 MBPs with the same capacity/number of NAND chips
from the same vendor?
Exactly my thoughts. What's weirder is that the 512Gb 13" seems to reach the advertised 2500Mb/s (at least that's what I gather from the reviews).
Is there a way of knowing for sure the number of NAND chips on my machine?
Probably, though the question would be is there a difference depending on capacity.
The answer seems to be yes, unfortunately. It appears that for some (bizarre) reason the 13" gets two 256Gb in RAID0 and the 15" gets only one with half the speed (which would be total nonsense considering not only that the 13" is cheaper, but also that the 15" would be 'more pro').
I wouldn't mind it being slower as long as it were clearly stated on Apple Store's configuration page (I even looked for it before making my purchase! The only thing mentioned is the difference in capacity, never speed).
All in all, after holding back my purchase by the whole throttling fiasco, this is very frustrating. If I had been informed in advance I'd have gotten the 1Tb model, it's not only more storage it's twice as fast (apparently).
The 512GB i9 version we're publishing the review for shortly exceeds 2500 MB/sec.
How long have you had the machine? There is a pile of indexing the drive after a new user migrates data that can run for days, and this will impact SSD benchmarks.
I got it five days ago and I didn't migrate any data, plus I don't see mds or any suspicious processes running while testing the SSD. I even tried turning off FileVault (and saw no difference). I've even tried the test on different days before posting here.
I would immediately buy this model if offered with function keys (and a $200 discount). Now I am more reluctant to do that. Almost all programs I need to use frequently have no or very limited support for the touchbar, while I really need a decent ESC key
While I'm not super-excited about it, out of curiosity, have you actually used the Touch Bar for more than a few minutes?
I would immediately buy this model if offered with function keys (and a $200 discount). Now I am more reluctant to do that. Almost all programs I need to use frequently have no or very limited support for the touchbar, while I really need a decent ESC key
While I'm not super-excited about it, out of curiosity, have you actually used the Touch Bar for more than a few minutes?
Yes, I did. The graphical designers in the company like it because the graphical tools support it. I find it rather useless for development activities I am doing.
- All command line activities in the Linux VM's I am running, have not much help from the touchbar.
- The coding utilities I am using (mainly PyCharm) have some limited support for touchbar, but use function keys everywhere, so it is actually counterproductive to have a touchbar.
- All my apps are heavily using the esc key. On a normal keyboard I can hit the esc key blindly without error, on a touchbar model I can't.
A big improvement would be a slightly shorter touchbar and next to it an real esc key
This just in: my friend who got the 15" i9 variant with 32Gb of RAM and 512Gb SSD is also seeing ~1700Mb/s.
(Maybe it's just me, but this seems like a hugely important info)
The Blackmagic Disk Speed Test is designed to test/stress the SSD device
the same way video does. It does not seem to really indicate exactly
what tests it is running (compressed data vs non-compressed, sequential
read/write vs random access reading and writing, block size of the data
etc). SSD specifications (i.e. package specs) only tend to publish the
specs that you will get the highest rating from... which I believe is
probably sequential read and then sequential write-only.
I used Blackmagic because that's what I saw on most (if not all) of the reviews out there. If anyone has any other (free/open) suggestions, or other ways of testing the SSD speed, I'd be glad to try them.
Regardless of what the type of test Blackmagic does, shouldn’t it be
consistent for all 2018 MBPs with the same capacity/number of NAND chips
from the same vendor?
Exactly my thoughts. What's weirder is that the 512Gb 13" seems to reach the advertised 2500Mb/s (at least that's what I gather from the reviews).
Is there a way of knowing for sure the number of NAND chips on my machine?
Probably, though the question would be is there a difference depending on capacity.
The answer seems to be yes, unfortunately. It appears that for some (bizarre) reason the 13" gets two 256Gb in RAID0 and the 15" gets only one with half the speed (which would be total nonsense considering not only that the 13" is cheaper, but also that the 15" would be 'more pro').
I wouldn't mind it being slower as long as it were clearly stated on Apple Store's configuration page (I even looked for it before making my purchase! The only thing mentioned is the difference in capacity, never speed).
All in all, after holding back my purchase by the whole throttling fiasco, this is very frustrating. If I had been informed in advance I'd have gotten the 1Tb model, it's not only more storage it's twice as fast (apparently).
The 512GB i9 version we're publishing the review for shortly exceeds 2500 MB/sec.
How long have you had the machine? There is a pile of indexing the drive after a new user migrates data that can run for days, and this will impact SSD benchmarks.
I got it five days ago and I didn't migrate any data, plus I don't see mds or any suspicious processes running while testing the SSD. I even tried turning off FileVault (and saw no difference). I've even tried the test on different days before posting here.
Comments
Dissent is fine. Being a jackass and/or violating the commenting guidelines is not. I'm done giving warnings.
Edit: That didn't take long. Banned.
Vadim's measurements were probably enough for me to narrow down my choice, thanks! A few percent performance gain, even with the larger cache? Not so tempting.
I used the term "battery life", but meant to ask how long the thing runs on battery with minimal load. Have you thought about trying to figure out how much extra power the non-LP RAM consumes? If only to guess at the decisions Apple might have been making putting the 32G option off.
And don't get me started on the keyboard. Apple knows better and could do better with the keyboard. It almost seems like intentional self-sabotage, given that Apple can't outright kill the MacBook Pro line. Instead, take a key aspect of the laptop -- the keyboard -- and make it uncomfortable.
The Blackmagic Disk Speed Test is designed to test/stress the SSD device the same way video does. It does not seem to really indicate exactly what tests it is running (compressed data vs non-compressed, sequential read/write vs random access reading and writing, block size of the data etc). SSD specifications (i.e. package specs) only tend to publish the specs that you will get the highest rating from... which I believe is probably sequential read and then sequential write-only.
I believe the highest Boost clock we saw was 4.2GHz for the i9 (all 6 running, of course)
The issue is that people are fooled by that incredibly high Boost clock speed.
The conditions have to be extremely specific for one core to reach that high.
We'll have more information on this when our 15" i9 review goes live tomorrow. Make sure to check it out.
Same keyboard shortcomings improvisingly repaired, same lack of MagSafe, same functionality - struggling to keep up with the competition RAM & graphics wise.
OK, the i9 is a beast but it has to be seen if the less than average cooling support system can handle it
Okay.
The 512GB i9 version we're publishing the review for shortly exceeds 2500 MB/sec.
How long have you had the machine? There is a pile of indexing the drive after a new user migrates data that can run for days, and this will impact SSD benchmarks.
I have specific reasons for not wanting to use FaceID but would like a quick method of logging in.
While I'm not super-excited about it, out of curiosity, have you actually used the Touch Bar for more than a few minutes?