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.
Mac Studio Performance - Disappointing News
- Thread starter CodeSpyder
- Start date
- Sort by reaction score
It might if the benefit they actually provided approached the benefit they could provide. Based on his testing it appears Apple, for unknown reasons, has prevented the GPUs from delivering the performance we would expect them to deliver.
I can assure you there’s a significant demographic of users that wanted to hear the Ultra’s 64x GPU scaled approximately double the Max’s 32x GPU (or more accurately, in a more linear way per core). Even if you leave specific use cases aside, why would anyone want to hear otherwise?
Last edited:
They wanted to hear that 8 cores was all you needed. For the M1 they wanted to hear that 8gb was the new 16gb.
If that's what they wanted him to say then I think they were disappointed as I didn't see any such statements in the Studio review I watched. Or did he make, and are you referencing, videos of the original M1 systems where he made such claims?
16 more cores. That's 33% more than the base model. Like most things in computing there are diminishing returns. The price is what it is--without knowing the binning fractions of units that have all 64 cores fully-functioning none of us can speculate on that price and what that means to Apple. Buy it if you think its worth it, don't if you don't. Pretty simple. I think Max's tests were poorly constructed to really test the systems in *real-world* daily use, and on top of that it's early days and much software optimization has yet to occur.
I can’t tell if you’re being intentionally obtuse or trolling. “telling people the i9 isn’t worth it vs the i7 in the imac 2020” is a widely held opinion and backed up by actual benchmarks.
The last iMac CPU upgrade that was notably worth it was the 2017 to the i7 7700K. After that on the 2019 and 2020 had questionable price-to-performance ratio for the i9 upgrades concerning single core. With multi-core it is more linear though.
So yes, I don’t think this example qualifies has a criticism against Max Tech. But I can agree sometimes they make generalizations on disciplines they clearly are not familiar with, especially Lightroom Classic, like lol any seasoned user could make a 16GB machine grind to a halt within 30 seconds.
So yes, I don’t think this example qualifies has a criticism against Max Tech. But I can agree sometimes they make generalizations on disciplines they clearly are not familiar with, especially Lightroom Classic, like lol any seasoned user could make a 16GB machine grind to a halt within 30 seconds.
The Geekbench multi-core benchmarks for the i9 are about 10% higher than the benchmarks for the i7. That said, people don't generally buy their computers to run benchmarks.
In the 2020, the i9 is a 10 core CPU and the i7 an 8 core. In this case the multi-core benchmark is the most relevant. Single core was still higher though.
I am referring to earlier videos but my comment is more about that channels clickbait strategy. They do have useful information but they editorialize too much.
Yes there are diminishing returns but the 64 GPU upgrade seems to be an extreme example. I think in all but one of his tests it was his opinion that the low single digit, if any, returns did not justify the price. It's $1,000 just for the 64 core GPU upgrade.
I would like to hear what you think would be a better test. In fact I bet he would welcome such feedback.
Regarding the optimization after having been in this industry long enough I have learned you buy a system for what it can do today and not what it may (or is promised to) do in the future. I've fallen for that a few times. A reviewer needs to be able to base their recommendation on observed behavior and not theoretical future benefit.
When the Mac Studio was announced I thought to myself: Wow! Look at all the compute power it contains. After watching a number of reviews I concluded the real world performance does not match the paper specs in the way one would expect. It appears Apple is limiting the ability of the system to utilize its full potential. I don't know the reason and I wouldn't use that as a reason to avoid buying one. Buy it for what it is and can do for you. But it's good to know that the extra money, especially for the 64 core GPU upgrade, seems to only benefit, at least in the benchmarks I've seen, a sliver of buyers.
If there were any better examples of why Geekbench is a lousy benchmark I think the various Mac Studio reviews are it.
That would explain it. I didn't watch many YouTube videos on the original M1 release so I likely missed such comments. I did read, and laughed a lot at, comments on this site which were making said claims.
Have you watched any of his Mac Studio reviews? I found them very helpful. The one I really enjoyed was the comparison of the fully loaded Mac Studio Ultra with the 12900K PC with a 3090 GPU (Luke Miani just did a similar video). Despite the Mac Studios "issues" it was quite competitive with that system.
There are issues both with his methodology and his presentation style, although it's the hyperbole and bombastic style of the presentation that's most annoying. All of the SHOCKING and excess is just unnecessary and purely to draw clicks and ad revenue. As for the testing, it's very limited and not terribly useful for real-world applications. If you have any knowledge of the various products used, in many cases the "test" cases used are just simply insufficient to show any differences between the models, so it's not necessarily the case that the specs are useless, but more that the test isn't useful for highlighting any differences. I personally own three different Studios (one base Max, one base Ultra and one fully loaded Ultra) and did my own extensive testing focused on photography use, and can assure you his tests are completely inadequate for a someone making an assessment for that use. Meanwhile if you spend any time poking around into more specialized testing, you'll find the same.
His conclusions on throttling and his fixation with wattage are completely misguided; without understanding exactly how the code in use is working and what is going on with how the AS architecture divvies up workloads for the GPU, it's impossible to draw most of the conclusions he is making, and in particular the strange ones focused on wattage. He's in a rush to make himself seem relevant and in the process makes the mistake of speculating on things where he has a best a basic working knowledge, yet he leaps to conclusions that are quite a stretch. These days you can get away with such things, because everyone will just move on the latest video with big flashy titles and ALL CAPS hyperbole. What's lost is real insightful knowledge from an expert perspective.
I find nothing wrong with his presentation style (well, maybe he says insane an insane amount of time). IMO whether a viewer agrees with his style of not is subjective.
If you disagree with his test cases then what would you recommend? I think he's been very clear his conclusions are limited to what he's tested and that one should consider them in the context of the viewers use cases.
In his conclusions though he literally says he doesn't recommend a couple of the higher end Ultra builds, at all, for any video editing, based solely on a few export tests. I agree that viewers need to be aware of the context of any reviewer's opinions, but he can also stay away from giant blanket statements based on a few preliminary tests (of course that means he loses some click-bait, and it's hard to fault someone for playing the click-bait game when it's a click-bait world). He also plugs an as-yet-to-come, more in-depth video editing review (which might not even lead to the same conclusions as this superficial shoot out).
What I'd like to see is more real world observations, such as when video editing does the app start to feel sluggish once you have a bunch of GFX, streams, filters, etc., in the timeline/project? How does the computer perform when you are exporting and editing at the same time? Or when you are jumping between multiple open apps like PS, AE and your NLE? I know they did the 8K export test, but what's the experience like of editing it? Can you zoom around the timeline w/o lag? Does it still playback smooth as glass once you start adding graphic overlays and color correction? I know that takes more time/effort than just running benchmarks/speed tests, but it also gives a better picture of the user experience and how the app actually performs vs just how fast it can export/render.
Doing a computer review mainly based on export, render times is like a doing a car review based solely on the 0-60 time and ignoring handling, cornering, braking, road noise, ride comfort... basically the entire car driving experience.
His recommendations are based on the tests he performed which are clearly stated to the viewer. The viewer has to decide for themselves whether the results are applicable (or not) to their situation. What he tests is clearly stated with nothing hidden and conclusions based on his opinions are clearly noted.
I'd like to know how Handbrake encoding performs but he didn't do it. Such is life.
Then by all means perform your own review and publish it. None of what you said warrants the criticism of what he has provided. If you don't find it of value then that's fine, however I found it valuable (despite one of my use cases not being covered). What I didn't find was that it was misleading.
Max? Is this you? 😂
Honestly, plastering terms like "TRUTH" and "Apple Slowed it Down" are misleading, period. He posts stuff he fully knows is just click-bait to get discussion and commentary. Telling people to publish their own reviews is a ridiculous response to the arguments being presented and a deflection from the issues at hand.
The reality is that he is trumpeting his own views as "TRUTH" and not as opinions, small side comments to the contrary notwithstanding. It's a bit like all of the opinion shows masking as "News" on television/online today. Many folks lack the skills to separate the two, and the people publishing that crap know it, but all they care about is the ad revenue.
Do you have a better explanation for the lack of GPU scaling? He provided his conclusions and how he arrived at those conclusions. Only a fool would consider them anything other than his opinion.
Can you provide specific examples where he has "trumpted" his own view as the absolute truth (as opposed to informed opinion)? Video link and time within the video please.
Last edited:
Only a fool would post such wild speculation on a topic on which they clearly know so little, and plaster it all over the Internet with headings like "THE TRUTH" in all caps.
Again, the burden of proof is on the one making the claims. Stop trying to deflect to criticism to those pointing out the obvious faults.
I agree. Since you're claiming he clearly knows so little about the subject matter the burden is on your shoulders. In my previous posts I asked you for specific examples and you provided none. I'm keeping an open mind but the only thing you've provided is your opinion.
You really don't get it, do you? Open mind, rigghhhhttttt. All you are is a Max apologist on this thread. Either you're Max, his brother, or his biggest fan. He's the one presenting himself as the reviewer. The burden of proof is on him, period.
I've provided my own analytical data on performance elsewhere, and when asked a legitimate request here by someone to run a real test, I have done so. What I won't answer is your logical non sequiturs and ad hominem attacks.