CanadaMark Offline Upload & Sell: Off
|
RoamingScott wrote:
Dramatically lower? The old M1 Ultra was at or near performance parity with the i9-12900 while using 25% of the power. The M2 chips are about 20% faster now at the same power consumption. That said, the Ultras are crazy overkill for 95% of the market, but the performance parity IS there between both ecosystems.
I get why most long term Windows/Intel users are wary of Apple given their history of hyperbolic marketing, but Apple Silicon is, again, the best bang for the buck you can buy for personal computing right now. The unified memory is extremely fast, and all of the fastest numbers I've seen for Lightroom 12.4 are coming from Macs.
IMO anyone that hasn't owned an M1/M2 mac is disqualified from passing judgement on it, especially in the realm of post processing. ...Show more →
Yes, dramatically lower, but there are outliers, for example M silicon does phenomenally well with video exports in certain software like Resolve. You can look at most objective benchmarks, including Apple's favorite benchmark like Geekbench ( ignores thermals, and heavily favors the workloads Apple silicon excels in), and the M2 Max does not fare well against the top end Intel silicon beyond single threaded performance. Blender, 3D Mark, Cinebench, etc. all heavily favor higher end Intel silicon and/or Nvidia 4000 series GPUs. In the case of the 4080 and 4090 they actually have dual encoders which cut video encoding times roughly in half. As soon as you introduce GPU-heavy tasks, obviously it's going to depend on what specific GPU you are comparing it to, but the GPUs in Apple's silicon simply cannot compete with something like mid-high tier 4000 series GPU with more VRAM and getting fed many times the wattage. It's also not designed to compete with those, so it shouldn't necessarily be viewed as Apple 'losing'. Just different goals.
My wife has a M1 Macbook Pro and I would say I recommend about 50/50 Macbooks and Windows machines to my friends/family/coworkers that come to me for advice. I have spent many years working in the industry as well. As I keep saying, it just depends on EXACTLY what you are using the machine for, it's really difficult to make broad generalizations such as X CPU outperforms Y CPU - ok, in what specific tasks? at what power levels? in what chassis? in what software? It's not cut & dry. Different CPUs and GPUs also have different hardware acceleration capabilities which are leveraged differently by different software, and those can make an enormous difference as well, again, depending on specifically what you're doing. Comparing hardware is generally easy because we have objective benchmarks to look at. The tricky part is understanding things like thermals and power levels if trying to make direct comparisons.
The M silicon is generally designed for maximum performance on minimum wattage, and to maintain performance on battery power, which for some users is a huge benefit. Performance per watt is outstanding, and battery life can be very good. It's very good at that and efficiency is very much a strong point. Most people don't need crazy amounts of power. However if you want the fastest possible machine, M silicon is not that, and that doesn't mean it's at all bad, it's just optimized for a different purpose.
In Lightroom, most tasks (like moving sliders) are single threaded, and M silicon is plenty good at single threaded tasks. Hobbyist photo editing really isn't that taxing on any system in general. It's more when you get into heavily multi-threaded or GPU-heavy tasks, and higher power levels that you will see larger differences, and that extends to software far beyond Lightroom. As a Lightroom editing machine for a typical user, M silicon is an excellent choice, particularly if you edit a lot on battery power.
Just as an example, here are some M2 Max benchmarks run by an Apple-centric website against a very powerful windows laptop:



They don't have charts but here is a quote from the same source:
"In Premiere Pro, the M2 Max MBP managed to very slightly beat out the HX on rendering a 30-minute 4K 60FPS 10-bit movie but fell substantially behind the MSI in exporting (12 and a half minutes versus 19 and a half minutes).
As for various games compatible with both systems, as you might expect, the RTX 4090 is optimized for gaming demands and handily beats the M2 Max's 32-core on-chip GPU."
Here is Cinebench and Geekbench data based on average user-submitted data (not one specific model vs another):

I know I sound like a broken record, but it very much depends on what you use the computer for. *Most* users, in general, would be happy with any decent laptop from either camp and probably never even know the difference. However if you are after maximum performance for a certain use case, then there are some major differences depending on exactly what you're doing and depending on your usage environment. Higher-end Windows hardware is overall clearly faster but you pay for it in size (usually) and wattage (almost always). You also usually need to be on wall power to get maximum performance (there are exceptions here too, but not in the highest performance tiers). Not everyone wants to make those compromises, and that's why both Macbooks and Windows laptops are both excellent choices depending on the usage case.
|