Hacker News Reader

Comment by rStar | Hacker News Reader

16/512gb on my m1 mini is a great computer, but it’s not as ‘snappy’ as i thought it would be from all of the reviews. i run a lot of adobe.

View on HN

Lightroom Classic (latest version) is unpleasant on my wife's M1 and even crashes. Last night I moved it over to my gaming PC which is pretty old, but performs much better.

It ran on her 2012 Intel MBP and whatever dayjob MBP I had in the past.

May I ask why you purchased the M1 model?

To replace the 2012 which was mostly OK, but had blown speakers.

I still have an Intel MBP provided by my employer. I don't keep a personal laptop, it would just gather dust 99% of the time.

First, thank you for your reply. I am curious, after this experience, would your next machine be also an Apple with M1 (or the next iteration of it), assuming nothing else goes wrong?

Probably, this is my wife's computer.

We aren't attached to Apple, but finding build quality, battery life, high quality screen and likely to last 10 years is hard elsewhere.

Thanks for the insight, I hope it works out for you all.

i’m hoping adobe has yet to fully optimize for the new architecture, and i’m also hoping they care to do so

It could also be that x86 has better SIMD support by a considerable margin, which can make repetitive memory access/serialized workloads a lot faster. I'm not super well intimated with those extensions, but I know that the NEON SIMD implementation in ARMv8 leaves quite a bit to be desired. It's a tricky situation, and one I don't see resolving in a nice clean way. It's stuff like this that makes me hopeful for RISC-V though, where we could theoretically have our cake and eat it too, with dynamic instruction pipelines and incredibly low power usage. Only time will tell, I suppose.

Throughput is not an issue on the M1, with 4x 128-bit SIMD units.

Neon is certainly not a bad SIMD ISA, it's a quite orthogonal one.

You also have the AMX extension at hand, which is more special purpose but allow to deliver very high throughput. (on a regular M1: 350Gflops DGEMM, 1.2Tflops SGEMM, without leveraging anything other than the CPU)

NEON can work pretty well and has been used in real time tasks on mobile for a long time. Example operations:

https://github.com/thenifty/neon-guide/

LR CC runs much much better on my M1 MBA than my prior 2017 MBP. I wonder if Adobe is just dragging their feet on updating classic since they want everyone moved to CC?

Reply on news.ycombinator.com