• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

ARM is coming.

Soldato
Joined
6 Oct 2009
Posts
4,014
Location
London

Wondering how long it will take? It is crazy the power usage difference. Like when will I have to go to ARM in my PC? It should not take long IMO.

Death of x86?!

ARM is already here, actually has been for a while and is gaining ground in the server market as well. x86 will be fine too, there's no reason for it to die.

Apple's migration to their own ARM chips has been very successful, and we'll see more ARM in PCs outside Apple in the coming years but there's no reason to think x86 will die. Nvidia is also interested in this and will likely make some moves as they've already done that in the server market.

Universal binaries and multi-arch design of software has been around in macOS and Linux for a good while now and will be the future of Windows too, to the end consumer it just wouldn't matter and it will all be seamless.
 
Last edited:
Soldato
Joined
1 Feb 2006
Posts
3,415
MS can do a x86 emulator that only uses the ISA instruction that the copyright/patents have timed out for. Last time I read about this (5+ years ago), they had a fully 386 one so it must be a lot better by now. Software that uses the newest instructions will be the biggest problem as Intel will not let them be emulated. MS should go RISC V, if Windows for ARM works out, ARM is the new Intel and the cycle repeats.
 
Soldato
Joined
28 May 2007
Posts
18,385
MS can do a x86 emulator that only uses the ISA instruction that the copyright/patents have timed out for. Last time I read about this (5+ years ago), they had a fully 386 one so it must be a lot better by now. Software that uses the newest instructions will be the biggest problem as Intel will not let them be emulated. MS should go RISC V, if Windows for ARM works out, ARM is the new Intel and the cycle repeats.

The biggest problem is the use of dedicated accelerators and the non disclosed/documented/blackbox features. Although Intel have been more willing to give people a peak behind the curtain… I wonder if Intel would be more amenable to Open Common Socket proposal today.
 
Associate
Joined
8 Oct 2020
Posts
2,355
So what will you run on it with no backwards compatibility?
I think he’s referring to the trash legacy desktop apps that multibillion dollar companies still run their accounting on.

Any well developed and actively maintained app, especially anything built in recent years, shouldn’t have an issue with an ARM build. There would still be a compatibility layer for x86 but again, for modern apps.
 
Associate
Joined
3 May 2021
Posts
1,228
Location
Italy
I think he’s referring to the trash legacy desktop apps that multibillion dollar companies still run their accounting on.

Any well developed and actively maintained app, especially anything built in recent years, shouldn’t have an issue with an ARM build. There would still be a compatibility layer for x86 but again, for modern apps.
What's a well developed and maintained app?
In my experience most project gets progressively worse in quality as they scale...
 
Associate
Joined
8 Oct 2020
Posts
2,355
What's a well developed and maintained app?
In my experience most project gets progressively worse in quality as they scale...
Maybe "well developed and maintained" is a bit vague as software is very much a spectrum, but any common and widely used app e.g. MS Teams, Discord, Steam, etc. shouldn't require significant rework to have it available on ARM, whereas I wouldn't expect or care about some dodgy exe that a dude still uses to read the ECU on a 30 year old tractor.
 
Don
Joined
19 May 2012
Posts
17,281
Location
Spalding, Lincolnshire
Oh dear...

Both reported numbers that were nowhere close to what Qualcomm promised. How not close? Above 50% this time but one used the term ‘Celeron’ to describe performance. The claims of better than Apple’s Rosetta 2 x86 emulation are clearly not real on what is probably the release hardware and software. Actually the silicon emulation may be better but everything else is unquestionably not.

Two major OEMs with serious engineering capabilities are strong evidence but not proof. Here is where we will blur things out more than we like, sorry about that. A while back we were digging on performance or lack thereof in preparation for the promised briefings. A deep source at Qualcomm told us that the benchmarks were cheats, told us how they were cooked, and told us that Qualcomm was well aware of it. This same technique meant the numbers looked far better than they could be on a non-trivial set of tests. Ironically some other benchmarks could have looked much better than those presented if Qualcomm adequately disclosed their testing details. Lose some, lose some, but no winning here.


 
Associate
Joined
8 Oct 2020
Posts
2,355
I fully believe that the new Qualcomm chip will be a huge leap forward for non-Apple ARM, but I've set my expectations well below the advertised benchmarks.

Performance-per-watt matters a lot, and they're running these chips with more juice than the Apple processors. There's also no way the software will be anywhere near what Apple have already had running live for several years now.
 
Soldato
Joined
25 Sep 2009
Posts
9,664
Location
Billericay, UK
Oh dear...




I haven't read one of his articles for years, tbh I thought he was dead.

Typical Charlie Demerjian article, lot of words and very little detail but I get his point. I suspect a lot of that is to do with the emulation and state of Widows ARM, first gen hardware is never optimised but lets see if Microsoft can pull a rabbit out the bag with some software updates.
 
Last edited:
Back
Top Bottom