Actually thanks for pointing it out and brings it to attention to many more, when I saw the Mac Pro and iMac Pro SSD speed I was underwhelmed. Hopefully more people are aware of that.
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.
Apple's 2019 256GB MacBook Air Includes Slower SSD Than 2018 Model
- Thread starter MacRumors
- Start date
- Sort by reaction score
Actually thanks for pointing it out and brings it to attention to many more, when I saw the Mac Pro and iMac Pro SSD speed I was underwhelmed. Hopefully more people are aware of that.
you do know that these benchmarks are like twice as fast as the Dell XPS 13? f*****ing ridiculous!
Thanks for the explanation. I just bought a 2019 MBA and was getting nervous about this headline. Now I'm pretty sure I won't notice a difference -- except a 1-3 seconds more to wait when opening or moving the pretty occasional excessively large file. So why not the price drop! Still going to be a significant upgrade from my 2014 MBA.
which? I've got a couple Dell XPS13" devices floating around the office. Gave one to our dta guy. benchmarked the storage and it was faster than 1000mbit/s.
During the first couple years the XPS 13" was shipping with SATA based storage, hence slower drive speeds.
Current ones ship with NVME and are swappable.
iMac Pro:
- 3.3GB/s write performance
- 2.8GB/s read performance
Do you even know the source fro your "revelation"?
[doublepost=1563214751][/doublepost]
worse SSD, but still pretty impressive!
This is Apple's way. They will reinstate the SSD speed in the next iteration and then talk about how much they've improved.
True, Dells are all over the map, depending upon the SSD you get, the stock in a $1000 Dell is pretty damn slow. Google to check. One thing with Dell, they don't guarantee components or speeds on their SSds cause they like to swap in cheaper drives if the become available. Buyers recommend you get the cheapest one, then swap it out for the Samsung 970 after delivery. But some are faster than others.
This is from https://www.laptopmag.com/reviews/laptops/dell-xps-15-2019. and references the XPS 15 - which should be no slower than the 13, not the 13
Dell XPS 15 (2019). 508
Apple MacBook Pro (15-inch, 2019)
write (Black Magic) 2610.8
read (Black Magic) 2612.1
read (Black Magic) 2612.1
HP Spectre x360 (15-inch, OLED) 424
Category Average (as of 06/30/19) 655.54
And to keep Apples to Dells, the article specifically mentioned Blackmagic, even benchmark software can yield different results.
Now I'm genuinely curious. If I can "steal" that laptop back from the data guy, I'm going to benchmark it to see. Genuinely curious.
I benchmarked my Razer Blade Stealth (2017 model, i7-7500u with 256GB SSD) and it's a non-branded samsung NVME and achieved the 3000mbits read and 2200mbit write.
the Lenovo work laptop I have is also an unbranded NVME, just shy of 2900 MBIT read and 2500MBIT writes. But I've swapped that one out with a brand 970 pro
FYI, this site visitors/members are from all corners of the world.
Plenty of places on MR where you can find this, seems like you look in the wrong places.
Not the quality of Apple's SSD's, this has been discussed ad nausea.
Let me educate you, those SSD's are TLC while Apple's ssd's are MLC, big difference.
As above.
You’re not wrong. But the difference in customer experience is marginal at best.
What a cheesy back handed thing to do when Apple touts it as getting MORE for less. 
No offense, either, dude, but expecting a Windows VM to be performant on a dual-core/four thread 5w TDP ULV CPU with 8GB and then complaining that because Apple is using a slightly slower SSD, that that is the cause of it running slowly is a bit like lifting a Ferrari’s suspension to monster truck levels and then being upset when it fails to lap the Nurburgring in 7:23.
EDIT: Correction...it's a 7w TDP on that dual-core/four thread ULV CPU, not 5w TDP. Just want to get that in there for the record that I recognized my error and the additional 2w TDP of headroom for which to bust a cap on the Windows VM.
Last edited:
What would you buy? considering there's really NOT faster storage available. even the MacBook Pros which are considered the fastest are using the same basic tech and speeds are comparable.
There's no faster consumer grade drives without moving into SAS/SCSI, and even those don't necessarily beat out NVME drives.
But the new MB Air is still faster than what you benchmarked...
Or could it be that your benchmarks are off by a factor of 8? Bit... Byte... anyone?
But the new MB Air is still faster than what you benchmarked...
Or could it be that your benchmarks are off by a factor of 8? Bit... Byte... anyone?
Thanks, I was wrong with which speed acronym I was using.
Byte. 3200MegaBytes per second.
I frequently flip those two around, by accident as I'm often flipping around depending on what talking about. it's important to be clear though.
Useless Anecdote Time: For the last few months CTO has been crying foul that we're being bottlenecked by our ISP and that we're only getting 1 megabyte per second over a particular pipe we have. I tried to get him to confirm with the ISP exactly what we paid for... He claimed over and over again. it was 10 Megabyte per second and we must yell at them for throttling us.
I pulled the invoice. it was 10megaBITS per second.
it's a significant difference
I need to remind myself to pay closer attention and use the correct measurement.
I think you're getting confused with MacRumors readers.
All about squeezing as much as they can down to the cents in places where the average consumer might not notice when purchasing. And still, what about the lower quality panel and almost unusable front camera.
This is indeed the case. As a developer, I compile very large C++ code. I went to the extent to using a RAM disk instead of an SSD, hoping it would be significantly faster to write there. It made exactly 0 seconds difference in the overall build time. What mattered was the number of parallel threads, but the speed of the storage didn't make any measurable difference. Especially because the system caches the files, and with a big enough memory you never have to wait for writing the output to the SSD, because it just fits in the memory.
This is also true for exporting photos and videos. The speed of the SSD is completely negligible compared to how long it takes to render the output. When you write a file to the disk, it's instantly written into the RAM first. Applications almost NEVER have to wait for a file to be written to the disk, because it happens in the background.
Disk speed only matters if you're copying terabytes of raw data from one drive to another. It doesn't matter during "normal" application mode.
Your post makes no sense. If the strategy is better for the stock price, why fire the CEO?