Mac New Game Porting Toolkit is Wine

I am personally more interested in seeing what the performance hit turning RT on would be for Mac hardware. The only game I have that requires RT is Metro Exodus Enhanced Edition.
 
GPTK 2.0

Gokhan Avkarogullari (@gavkar@mastodon.gamedev.place)

Game Porting Toolkit 2 is out https://developer.apple.com/games/game-porting-toolkit/ #wwdc24
mastodon.gamedev.place mastodon.gamedev.place

developer.apple.com

Game Porting Toolkit 2.0 - Games - Apple Developer

The Game Porting Toolkit makes it easier than ever to bring games to the unified gaming platform across Mac, iPad, and iPhone.
developer.apple.com developer.apple.com

AVX2 support, ray tracing, etc …

1718056228401.png
 
forums.macrumors.com

M3 vs M4

As usual Vadim appears to be wrong about technical matters. Not sure where he got that from. Also ARMv9 AVX2 support what's he on about? Is he confusing AVX2 with SVE2? Either way it doesn't matter since the M4 doesn't have SVE2 either. Its vector processors are still NEONs. Maybe he thinks...
forums.macrumors.com forums.macrumors.com

In further adventures with MaxTech misunderstanding tech.

EDIT: And courtesy of Pressure at MR for confirmation Ratchett and Clank working on the M1:



EDIT2: and MaxTech has deleted the original post, unclear if he issued a correction since I don't have Twitter to see.
 
Last edited:
dada_dave said:
forums.macrumors.com

M3 vs M4

As usual Vadim appears to be wrong about technical matters. Not sure where he got that from. Also ARMv9 AVX2 support what's he on about? Is he confusing AVX2 with SVE2? Either way it doesn't matter since the M4 doesn't have SVE2 either. Its vector processors are still NEONs. Maybe he thinks...
forums.macrumors.com forums.macrumors.com

In further adventures with MaxTech misunderstanding tech.

EDIT: And courtesy of Pressure at MR for confirmation Ratchett and Clank working on the M1:



EDIT2: and MaxTech has deleted the original post, unclear if he issued a correction since I don't have Twitter to see.
Click to expand...

NIce! I very much look forward to Reddit telling me that gaming on macOS was, is and will always be crap. No matter what happens. It’s 100% game coverage or nothing in between.
 
Jimmyjames said:
NIce! I very much look forward to Reddit telling me that gaming on macOS was, is and will always be crap. No matter what happens. It’s 100% game coverage or nothing in between.
Click to expand...
I don't know if you saw the original Vadim post that spurred my reply above, but oh boy ... it was bad.
 
Jimmyjames said:
He deleted it. My reply was nothing nasty.
View attachment 29880
Click to expand...
Fair enough. You know ... I'm possibly being a bit harsh in the MR thread, but as @leman said he has a big audience and putting stuff like this out there is just ... I suppose it is partially disappointment as he has consistently failed to learn. He clearly wants to talk about this stuff in more depth and it would be good for the Mac community to have a(n english speaking) popular Youtube source who could reliably break down more technical aspects of performance and power and architecture* etc ... I've just given up hope that it'll be him.

But also not correcting is not okay. Can't just delete stuff.

EDIT: *I'm not asking for this mythical popular TerchTuber to be at the level of Dougall or chipsandcheese or Andrei, but just something better. There is Geerkwan, which is fine for some of us, but obviously that's going to have its limits in terms of reach for english speaking audiences.
 
Last edited:
dada_dave said:
Fair enough. You know ... I'm possibly being a bit harsh in the MR thread, but as @leman said he has a big audience and putting stuff like this out there is just ... I suppose it is partially disappointment as he has consistently failed to learn. He clearly wants to talk about this stuff in more depth and it would be good for the Mac community to have a(n english speaking) popular Youtube source who could reliably break down more technical aspects of performance and power and architecture etc ... I've just given up hope that it'll be him.

But also not correcting is not okay. Can't just delete stuff.
Click to expand...
I don’t think you’re being harsh tbh. He is always jumping to conclusions and dismissing those who disagree. He decides to pursue an agenda and runs with it no matter what the truth is. I clearly remember the “M2 is a stop gap” he spread. He wouldn’t listen to anyone who disagreed.

Obviously the “M4 has AVX2“ was too far even for him.
 
Jimmyjames said:
I don’t think you’re being harsh tbh. He is always jumping to conclusions and dismissing those who disagree. He decides to pursue an agenda and runs with it no matter what the truth is. I clearly remember the “M2 is a stop gap” he spread. He wouldn’t listen to anyone who disagreed.

Obviously the “M4 has AVX2“ was too far even for him.
Click to expand...
Funnily enough the M3 turned out to be the stop gap! 🤪
 
Jimmyjames said:
NIce! I very much look forward to Reddit telling me that gaming on macOS was, is and will always be crap. No matter what happens. It’s 100% game coverage or nothing in between.
Click to expand...
If you are going to emulate/translate may as well do all of it, right?

Aside from anti cheat we are much closer to not needing Windows to game.
 
First performance numbers for AVX2 emulation:



I don’t have enough experience to know how reasonable these numbers are, if we should expect improvements, or how much game performance hinges on them.
 
mr_roboto said:
No link, but there's an inherent slowness to LCD technology. To change the color of a pixel, first a different electrical field is applied to the pixel site, then the liquid crystal material needs some time to reorient itself to the new field. There are various techniques for making this physical movement faster, but this is not a technology where you can have your cake and eat it too. Apple has traditionally prioritized color spectrum, accuracy, and off-axis viewing in their LCD panel designs, so they end up with relatively poor switching speeds.
Click to expand...

I was wondering ... Notebookcheck has been measuring the display response times and the 14" M3 and M4 were awful, but the variance around the measurements (14" M1 and M2 are good by comparison, especially the M2) and the 16" is even much better. They say Apple's method of dimming the display, using Pulse Width Modulation, essentially turning the screen on and off very fast makes measurements difficult but the stand by their measurements.

www.notebookcheck.net

Apple MacBook Pro 16 2024 review - Enormous battery life and better performance of the M4 Pro

Notebookcheck has tested the new Apple MacBook Pro 16 with the M4 Pro SoC, 48 GB RAM and a matte nano-texture glass panel.
www.notebookcheck.net www.notebookcheck.net

I summarized the findings in the comments and will ad the links that I couldn't add there since I'm a guest poster here:

I know very little about displays, but the interesting thing about the NBC response time data is its variance over time.

I can't post the links but you can check the 14" 2021 M1 Pro laptop review: 40.4/58.4ms

14" M2 Pro laptop review: 26.4/35.2ms

14" M3 Max laptop review: 78.8/80.8ms

The M3 Pro 14" is even worse than the M4 Max 14" but otherwise the M1 and M2 models have much better response times and we have this 16" M4 display panel which looks more like the M1/M2 14" in response times. As far as I know, the fundamental display tech isn't supposed to have changed much (I believe the M4 display has an improvement in SDR brightness and obviously the matte option but that *shouldn't* matter). I think Apple also has multiple display providers. So while typically Apple is known for exercising tight controls over its component suppliers maybe less so on display response time? and rather than a temporal change this represents a difference between suppliers? Finally, maybe the test measurements on these panels is flawed. The article mentions the difficulty in accurately measuring response times given Apple's high PWM flickering and perhaps all the measured response times are simply inaccurate. Maybe someone else has other ideas or knows how or why the variance is so huge.
Click to expand...

Do you (or anyone else here) have any ideas of what might be going on here?
 
dada_dave said:
I was wondering ... Notebookcheck has been measuring the display response times and the 14" M3 and M4 were awful, but the variance around the measurements (14" M1 and M2 are good by comparison, especially the M2) and the 16" is even much better. They say Apple's method of dimming the display, using Pulse Width Modulation, essentially turning the screen on and off very fast makes measurements difficult but the stand by their measurements.

www.notebookcheck.net

Apple MacBook Pro 16 2024 review - Enormous battery life and better performance of the M4 Pro

Notebookcheck has tested the new Apple MacBook Pro 16 with the M4 Pro SoC, 48 GB RAM and a matte nano-texture glass panel.
www.notebookcheck.net www.notebookcheck.net

I summarized the findings in the comments and will ad the links that I couldn't add there since I'm a guest poster here:



Do you (or anyone else here) have any ideas of what might be going on here?
Click to expand...
No idea unfortunately, but can’t wait for Tandem Oled to arrive on more devices.

Just skimming that notebookcheck article it seems like power usage is quite high. Is that concerning or is it just a case of Apple allowing power to rise to serve those demanding more performance?
 
Jimmyjames said:
No idea unfortunately, but can’t wait for Tandem Oled to arrive on more devices.
Click to expand...
Aye. Unfortunately Bloomberg has stated 2026 at the earliest and it's going to be expensive and Apple has gotten a bit gun shy because the tandem OLED on the iPad Pro screen didn't results in increased sales with the price increase.
Jimmyjames said:
Just skimming that notebookcheck article it seems like power usage is quite high. Is that concerning or is it just a case of Apple allowing power to rise to serve those demanding more performance?
Click to expand...
I'd say both. As I show here, efficiency took a big hit, especially for multicore. However, the multithreaded performance is huge compared to the M3 Pro and so efficiency (especially for the performance level) is still really good. I strongly suspect that not only does the clockspeed and number of P-cores (for multicore) play a role, but so does the 48GB of very fast LPDDR RAM and the 256bit bus. We should learn more when NBC tests the Max and hopefully a base M4. But the huge power draw of even the Pro chip under load is one of the reasons I would've ideally liked to see a bespoke configuration for the Pro that was somewhere in-between my M3 Pro estimate and the M4 Pro - 8+6 with a 192bit bus + the new RAM speed, but with the chopped die approach that isn't possible (unless Apple made the dies asymmetric wrt bus width and added a second P-core cluster onto the added bit for the Max, but Apple's never done that and I don't know if there are any technical limitations that would make that problematic).
 
No idea how much it will help with display efficiency or motion much but it seems the Mbp displays now use quantum dot technology.
1731631528205.png

Edit. Someone in the reply is saying pixel response is better in the M4. Obviously that isn’t really what Notebookcheck said.

1731631896341.png
 
Last edited:
Jimmyjames said:
No idea if it will help with display efficiency much but it seems the Mbp displays now use quantum dot technology.
View attachment 32774
Edit. Someone in the reply is saying pixel response is better in the M4. Obviously that isn’t really what Notebookcheck said.
Click to expand...

Quantum dots are neither quantum nor dots. I do not approve.
 
Back
Top