Discussion Future ARM Cortex + Neoverse µArchs Discussion

Page 15 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

hemedans

Senior member
Jan 31, 2015
239
123
116
IMHO it's all relative.

There's nothing you really need that performance for really on an Android device these days, as game devs/studios stopped even pretending to try and port recent console and PC games to the platform years ago, having written off its security problems in preventing side loading piracy (really Google's fault there, and no others).

AFAIK they still haven't even got round to porting GTA4 to Android, let alone GTA5 which is already pushing 11 years old.

WoA on the other hand is another matter - I'm sure binary translation for x86 Windows legacy apps and games will take all the IPC you can muster for anything newer than 5-10 years old.
Many games are playable Via Termux and box64 including GTA V, and in few months Cassia will be released, X86 translator for Arm which is more user friendly.

Bylaw did Demo Cassia playing GTA IV and Bioshock at high setting, so Soon average user will play these games in their Android phones.
 

soresu

Diamond Member
Dec 19, 2014
3,273
2,549
136
Cassia will be released, X86 translator for Arm which is more user friendly
Cassia is just the overall package of software with some extra UI bells and whistles.

FEX-emu would do the actual x86/64 -> ARM translation.

Think of Cassia like Valve's Proton for Android - everything you need in an easy to access package.
 

SarahKerrigan

Senior member
Oct 12, 2014
735
2,035
136
About where I expected it to be.

The V2 core is pretty fierce. GraphicsMagick numbers were interesting. In a different way, were the John the Ripper and Monero fiascos - though I suspect those two came down to a lack of hand-written assembly fastpaths.
 

moinmoin

Diamond Member
Jun 1, 2017
5,094
8,098
136
Is this the official Arm Ltd. thread now? They just had their earnings call, and they admitted that they essentially doubled licensing fees from v8 to v9. The expectation that this will slowly be reflected in revenue and profit in the coming years sent their stock soaring, still over 50% up over the past week.
 

FlameTail

Diamond Member
Dec 15, 2021
4,094
2,465
106
Is this the official Arm Ltd. thread now? They just had their earnings call, and they admitted that they essentially doubled licensing fees from v8 to v9. The expectation that this will slowly be reflected in revenue and profit in the coming years sent their stock soaring, still over 50% up over the past week.
Perhaps the thread title should be edited to such. Do we really need to have a separate thread for ARM Ltd and then ARM core designs?
 

Shivansps

Diamond Member
Sep 11, 2013
3,875
1,530
136
I was looking into the new Mali GPUs that ARM released, the G710, G720, G620, G510 and G310.

G310 and G510 are intended to replaced Bitfrost-based G31/G52, used in cheap socs. I would have no idea on how to calculate their performance, but the G31 and G52 are so slow that it dosent really matters.

G710 replaces the G610, and has a core configuration of 7 to 16 cores (G610 had 1 to 6), seems like a huge jump in perf, the G610MP4 already performs like a Vega 3 +/-, a G710 MP8 may potentially land in 5600G territory, a MP16 may even compite with RDNA3 APUs, it would depend on memory configuration i guess.

G720 is 6 to 9 cores, and G620 seems to be a G720 but limited to 1 to 5 cores.
 
Jul 27, 2020
20,419
14,087
146
So it was for the better that Nvidia-Arm acquisition never happened?
Considering the impact ARM has on the world of communication/social media, $40B was a gross undervaluation. Idiot Musk should've gotten ARM instead of Twitter.

Then I could have watched with glee with yummy popcorn how he pushed everyone to embrace RISC-V.
 

FlameTail

Diamond Member
Dec 15, 2021
4,094
2,465
106
If Nvidia starts making ARM based SoCs for PCs, come 2025....

I am looking forward to them making a powerful SoC for handheld consoles (Steam Deck like).

The ARM architecture provides the benefit of immense power efficiency, which is essential in a battery powered device like handheld consoles, and Nvidia is the leader in computer graphics...
 

moinmoin

Diamond Member
Jun 1, 2017
5,094
8,098
136
If Nvidia starts making ARM based SoCs for PCs, come 2025....

I am looking forward to them making a powerful SoC for handheld consoles (Steam Deck like).

The ARM architecture provides the benefit of immense power efficiency, which is essential in a battery powered device like handheld consoles, and Nvidia is the leader in computer graphics...
Why do you ignore Switch? Do you want such a system to run WoA so badly?
 

soresu

Diamond Member
Dec 19, 2014
3,273
2,549
136
The ARM architecture provides the benefit of immense power efficiency
Even Jim Keller has stated power efficiency is not really intrinsic to ISA.

ARM mostly started by aiming low and then slowly started picking all the low hanging fruit for higher IPC while focusing on keeping power/area as low as possible.

x86/Intel/AMD on the other hand started out with wall powered systems and just went crazy on getting higher performance with power and area guzzling IPC+clock improvement, and then slowly started leaning towards the lower power/thermals ARM neck of the woods as the consumer part of the industry started trending towards mobile/laptop form factors.

Computation in the consumer space has effectively acted like convergent evolution in biology.
 

Doug S

Platinum Member
Feb 8, 2020
2,833
4,819
136
Considering the impact ARM has on the world of communication/social media, $40B was a gross undervaluation. Idiot Musk should've gotten ARM instead of Twitter.

Then I could have watched with glee with yummy popcorn how he pushed everyone to embrace RISC-V.

"Impact" is not how investors value companies, profits are. ARM was overvalued at $40B, and is massively overvalued at $100B. They will need to demonstrate an ability to generate several billion dollars in profit every year, with reasonable prospects for near double digit growth in the short to medium term to justify a $100 billion valuation. That doesn't mean the price will go down - as Buffett says about short selling "the market can remain irrational longer than you can remain solvent" but investing in it is basically playing the hype train - relying on a greater fool to buy your shares at a higher price than you paid for them. It is not a proper investment where the company is seeing returns that justify its valuation, and the stockholders are compensated either via dividends or via buybacks that increase their relative share of the overall company's stock.

How is ARM going to be able to raise prices enough to justify that valuation, without inviting competition from RISC-V or other alternatives that may appear that do not charge any royalties at all? They said they were on track to "double" licensing revenue with ARMv9 (at least that's how I read their statements) but they need to way more than double their revenue to make the kind of profit they need to make. Maybe they were only talking about architectural licensing revenue, but they make so little from Apple and other architectural licensees that doubling it isn't getting them too far.

Most of their money comes from core licensing, so they'll need to way more than double their revenue from that source. That's also the market most likely to move to RISC-V, on the lower end at first (which is already starting to happen in the embedded market) and eventually to consumer facing gadgets like phones. Google has been working on making RISC-V a "first class citizen" Android port, so if there aren't already Android phones using RISC-V there soon will be. It'll start on the very low end (sub $50) sold in the third world where saving even a few pennies in core licensing matters, but move up from there.

Let's say someone invented a way to do fusion power in a form factor that could power a home, including heating, charging your EV, etc. (a "Mr Fusion") cheaply and decided they cared more about changing the world than becoming a trillionaire. So they charge $1 more than their cost of manufacture. The "impact" of that company is off the scale, it makes ARM (and everyone else) look like a company selling plastic dog poop gag gifts by comparison, but it will not deserve a valuation even as high as ARM, let alone as high as Nvidia or Apple, because it will not be producing billions in profits every year as far as the eye can see.
 

soresu

Diamond Member
Dec 19, 2014
3,273
2,549
136
Then I could have watched with glee with yummy popcorn how he pushed everyone to embrace RISC-V.
The ARM China, ARM/Qualcomm debacles + the ARM/nVidia acquisition possibility have been doing that pretty well already.

At this point I'd say that they are hoping for the best, but planning for the worst.
 

soresu

Diamond Member
Dec 19, 2014
3,273
2,549
136
Let's say someone invented a way to do fusion power in a form factor that could power a home, including heating, charging your EV, etc. (a "Mr Fusion") cheaply and decided they cared more about changing the world than becoming a trillionaire
Mr Fusion could literally fit in the back of the Delorean though.

Domestic power solutions (at least for those with a proper house) can be a little less volumetrically efficient than that.

That is why even though sodium ion based battery packs would have less volumetric and kinetic (weight/mass) energy efficiency they will probably overtake lithium ion in the home once they are scaled up, as sodium is potentially a cheaper material to source.

In a mobile setting weight/volume outweigh cost.

In a stationary setting that will reverse, or at least become significantly diminished.
 
Reactions: lightmanek
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |