M2 13-Inch MacBook Pro With 256GB SSD Appears Slower Than Equivalent M1 in Real-World Speed Tests

[visualseed]

They were thinking: "We looked at who actually bought the entry level first gen version of this product, and absolutely no one was doing a significant amount of exporting of Photoshop images."

Just a few years ago they sold a two-core i5 version of this machine with a single thunderbolt controller and storage nearly half as fast as what we are now calling slow. AND when they offered a performance model with an i7, more thunderbolt goodness with faster ram and storage hardly anyone bought it opting for the low-end model by a significant factor. At least now they get to offer the budget and performance model with the same motherboard and CPU only differentiated by ram and storage.

 

[lk400]

Gotta squeeze that penny for stockholders, nothing to see here.

PS: Kidding, slower means more "privacy".

There’s plenty of real world criticisms about this issue, many were covered in the other thread.

Why resort to these low effort, very very tired comments.

What’s next, you’re holding it wrong, innovative my ass?

 

[sevoneone]

View attachment 2024033
LOL!!! Pathetic.

Indeed because sure as heck looks like it is bottlenecking rendering and memory swap... lol. I'm not surprised, Ritche seems to just be dedicated to being a career Apple Apologist lately.

View attachment 2023996

Sorry, there is no way Apple didn't know this would be an issue. I am sure it had less to do with money and more to do with making sure they had enough NAND chips to keep the base models in stock, BUT it doesn't matter because they had to have known and it should have been disclosed. People would have still thrown a fit, but they would have known what they were getting into and had the option to buy 512. Now it is just going to be people returning their laptops in rage and demanding the 256 M2 Air be fixed before it is released, which is likely impossible...

Good thing they didn't release an M1Pro or M2 Mac mini like this. Lots of people buy Minis to use them as Render/processing nodes. If you think this is bad, that would have been some very tough damage control.

 

[MaxYuryev]

What bugs me is that the article headline uses the term “real world” when clearly the example is not “real world”, show me how browser performance, opening an excel file or such is impacted - that would be “real world” for that config.
If I were in the market for this model, base config - I wouldn’t buy it, let my wallet speak.

Did you watch the video? There’s a lot more in it than the article states. I do show browser performance among other things..

 

[falkon-engine]

Looks like something that will be fixed in an update. Bet on it.

That update will be called the m3 MacBook Pro.

 

[falkon-engine]

When photoshop, Lightroom, Final Cut Pro are my day to day tools - why do I have a 256GB drive and 8GB RAM? Is that the real world???

It’s slower than the predecessor with the same 8 GB and 256 GB.

 

[AF_APPLETALK]

What?

Dell is 2x faster than Apple 2 chip solution

This channel is also saying how the 3.5 Ghz clockspeed doesn't seem to be getting hit at all, and how the fans aren't even kicking in when the machine hits triple-digit temps (in C).

I have been supportive of this machine, but between skrimping on the SSD and the crappy thermals, I gotta rethink that.

Based on what we've seen here, I'm kind of thinking the M2 Air is not going to be as great as everyone is expecting. This feels like how lately things like cereal boxes have been getting smaller, but for the same price.

 

[jOlivier]

Since it has been confirmed that the 512GB models have two 256GB SSD chips and are delivering the expected high performance, the two most-likely options are:

1. Apple cannot get 128GB SSDs in sufficient quantity so they are using one 256GB SSD because they can get supply of those (you can see the empty second SSD location on the systemboard pictures);
2. Apple can get 128GB SSDs, but their purchase volume on 256GB SSDs means they are similar in price or cheaper than 2x128GB so they are using a single 256GB SSD to save money.

If point #2 was the reason, they would just set the base model to 512GBs instead of 256GBs, no?

I suspect Apple just made a mistake and will hopefully rectify the situation, either dropping the 256GB model, or going back to two 128GBs NANDs. Those who aren't happy with their machine should return it though. All in all, it's a bit of a shame that this happened seeing as it's the launch of the M2 and all...

 

[burgman]

Whether or not this actually contributes to the speed difference is just a theory at this point, lacking any other information. Wait and see if Apple has any software update to address it. Can't evaluate it at all until you know the software isn't getting in the way.

Yes, is this a real world issue, or just geek rage over a meaningless number? Because in Geekworld nanoseconds are a lifetime...

 

[M5RahuL]

Now... THAT takes courage!

 

[M5RahuL]

unlikely. you can’t add a 2nd ssd chip in an update.

Don't underestimate the power of the Reality Distortion Field on MR!!

 

[Cheffy Dave]

Benchmark testing has indicated that the 256GB variant of the 13-inch MacBook Pro with M2 chip offers slower SSD performance than its M1 equivalent, and now real-world stress testing by YouTuber Max Yuryev of Max Tech suggests that the 256GB SSD in the 13-inch MacBook Pro is also underperforming in day-to day-usage.

The M2 MacBook Pro with 256GB SSD and 8GB RAM was slower than the M1 MacBook Pro with 256GB SSD and 8GB RAM across multiple usage tests involving Photoshop, Lightroom, Final Cut Pro, multitasking, and file transfers. In a multitasking RAM test, the M1 consistently loads content faster with multiple apps open, and in a 50 image export test in Lightroom with apps open, the M1 was again quicker. It was able to export 50 images in 3 minutes and 36 seconds, while the M2 took 4 minutes and 12 seconds.

In these tests, the built-in 8GB unified memory of the MacBook Pro is being used by various processes, with the machine using the SSD for virtual memory. The virtual memory swapping results in slower system performance overall.

These results were consistent across all of the performance stress tests done by Max Tech, and benchmark tests conducted by Max Tech on Saturday demonstrated the same discrepancy. The M2 MacBook Pro's read speeds appear to be around 50 percent slower, while the write speeds appear to be around 30 percent slower.

• 13-inch MacBook Pro (M1/256GB) Read Speed: 2,900
  
• 13-inch MacBook Pro (M2/256GB) Read Speed: 1,446
  
• 13-inch MacBook Pro (M1/256GB) Write Speed: 2,215
  
• 13-inch MacBook Pro (M2/256GB) Write Speed: 1,463

Max Tech attributes this performance difference to Apple's choice of NAND flash storage. In the M2 MacBook Pro, there is a single 256GB NAND flash storage chip, while the M1 MacBook Pro has two NAND chips that are likely 128GB each. Multiple NAND chips allow for faster speeds in parallel, which could account for the M2's seemingly disappointing performance.

Slower SSD performance appears to be limited to the 256GB version of the 13-inch MacBook Pro, as higher capacity machines have not demonstrated the same issue. Potential MacBook Pro buyers should be aware of this performance problem as it could impact purchase choice.

It is not clear why Apple opted for a different NAND chip setup in the M2 MacBook Pro, and further testing is required to determine just what is going on. Apple has not responded to our requests for comment as of yet, but we will update this article if we hear back.

Article Link: M2 13-Inch MacBook Pro With 256GB SSD Appears Slower Than Equivalent M1 in Real-World Speed Tests

So how does this reflect on The MDA M2?

 

[jameslmoser]

No, at the moment that's a theory on why it's performing worse, because that could be why, lacking any other information. It doesn't guarantee it. All you know is that currently performing real world tests on the machines using the latest OS gives these kinds of results.

I wish I had a couple bucks for each time I've heard someone say "Can't be fixed, has to be hardware" when a new Mac is released. And then it's fixed.

How much you wanna bet me on it??

The fact they did that in the Mac PRO of all machines, and the fact they do the same thing in the smallest storage version of mac studio tells you all you need to know.

 

[mnsportsgeek]

This does not instill confidence in the m2 air.

 

[leicaman]

Considering every single SSD that is that size is slower than bigger versions of the same drive.

From every brand.

This is not news.

 

[HighDesert50]

One of Apple's suppliers is Kioxia (Toshiba). Kioxia suffered major production contamination earlier this year and that may be contributing to shortages and consequential compromises in hardware.

 

[jav6454]

So how does this reflect on The MDA M2?

Likely those models will also reflect such speed loss if Apple used just 1x 256GB chip.

 

[The Cappy]

Looks like something that will be fixed in an update. Bet on it.

If the issue is that the M1 used to RAM chips and the M2 used one, for less throughput, what's the update you have in mind?

 

[CWallace]

If point #2 was the reason, they would just set the base model to 512GBs instead of 256GBs, no?

Apple charges $200 for the 512GB storage configuration so making that the base storage (rather than 256GB) would have raised the price to $1499 from $1299.

If #2 is the reason (one 256GB module was cheaper than two 128GB modules), Apple would have done that to keep the base price of the M2 model at the same $1299 as the M1 model since they could have increased the price to $1399 to account for the higher component, production and shipment costs. So Apple clearly feels that the 13" MBP has to have a $1299 base price - it could be that this is the 13" MBP configuration they sell the most volume of and it is that popular because of that $1299 price.


Apple raised the base price of the M2 MacBook Air by $200 compared to the M1 MacBook Air, while still keeping the M1 MBA in the line-up at the same $999 price. It will be interesting to see what storage configuration Apple uses for the 256GB model. My gut tells me it might very will be the same 128GBx2 configuration and then we have to decide is it that way because Apple baked in the extra component costs in the $1199 price or is it because Apple was in fact supply-side constrained on 128GB SSD modules and had to delay the M2 Air's ship date to account for that?

An even nastier surprise might come if supply-chain constraints were the reason and later shipments of the M2 MacBook Pro 13" return to a dual 128GB configuration for the 256GB model.

I suspect Apple just made a mistake and will hopefully rectify the situation, either dropping the 256GB model, or going back to two 128GBs NANDs. Those who aren't happy with their machine should return it though. All in all, it's a bit of a shame that this happened seeing as it's the launch of the M2 and all...

Apple does not make mistakes of this nature.

The 13" M2 MBP has the storage configuration it has either due to cost constraints to maintain the $1299 base price or because there are not enough 128GB modules available (considering Apple is still using them in the M1 MacBook Air and possibly other products).

 

[jav6454]

If the issue is that the M1 used to RAM chips and the M2 used one, for less throughput, what's the update you have in mind?

No, RAM uses DRAM; Storage uses NAND.

 

[CWallace]

Considering every single SSD that is that size is slower than bigger versions of the same drive.

From every brand.

This is not news.

Indeed that is not news as the larger storage capacities (1TB and 2TB, especially) do gain from the higher cell count in the larger chips.

However, having a storage configuration with two storage chips that can be written to and read from in parallel will provide significantly more performance than what can be gained from a larger chip. For example, a 1TB storage configuration composed of two 512GB chips will outperform a 2TB configuration composed of a single 2TB chip even though the 2TB chip has some four-times the cells of each 512GB chip.

 

[jav6454]

Indeed that is not news as the larger storage capacities (1TB and 2TB, especially) do gain from the higher cell count in the larger chips.

However, having a storage configuration with two storage chips that can be written to and read from in parallel will provide significantly more performance than what can be gained from a larger chip. For example, a 1TB storage configuration composed of two 512GB chips will outperform a 2TB configuration composed of a single 2TB chip.

That is true, the end result is always to have the most chips possible. In this case 2, but it seems Apple nixed that idea.

 

[spartan1967]

What is so undependable about SSD R.W speeds, compared to only a few years back it only recently that we are seeing significant improvements. With any design that tends to allow a lot of latitude with how the laptop is configured its up to you to find out what best, not view the base model as all you need.

"...The M2 MacBook Pro with 256GB SSD and 8GB RAM was slower than the M1 MacBook Pro with 256GB SSD and 8GB RAM across multiple usage tests involving Photoshop, Lightroom, Final Cut Pro, multitasking, and file transfers..."

 

[jav6454]

"...The M2 MacBook Pro with 256GB SSD and 8GB RAM was slower than the M1 MacBook Pro with 256GB SSD and 8GB RAM across multiple usage tests involving Photoshop, Lightroom, Final Cut Pro, multitasking, and file transfers..."

Some people have selective hearing and reading.

 

[jdawgnoonan]

Well, if I was buying the lowest end MBP 13" I would take this as even Apple questioning why I was choosing it.