With the foundation of the M2 Ultra , bringing usMac Studio updatesand theMac Prowith it , Apple has complete its transition by from Intel and wrapped up the M2 crime syndicate of processors .
Now , all eyes are looking forwards to M3 . Reports say thatApple is already testing young laptops with the M3and the first possiblyarriving later this year . We do n’t know much about the M3 yet , except that it will likely be made using TSMC ’s 3 nm manufacture summons .
If we desire to have some idea of what to expect with the M3 , we can look at the phylogenesis of M - serial mainframe and how they relate to the A - series used in iPhones and iPads . Nobody really know what the M3 will bring until Apple tells us , but we can use the past to help predict the hereafter .
Updated 07/18/23 : A spreadsheet erroneousness caused some inaccurate Geekbench numbers . Those have been define .
How the M3 relates to the A16
When weestimated the public presentation of the M2 processorearly last year , we based our assumption on a simple concept : just as the M1 is essentially a big A14 with a few pinch , the M2 would be a expectant A15 . If you look at the public presentation difference between the A15 and A14 , you could figure the carrying into action gap between the M2 and M1 .
That turned out pretty well – our prediction were n’t far off . We believe the M3 will be based on the same architectural updates retrieve in the A16 , only again with more cores and a few other fitting . Is it comely to say the M3 will have the same performance gap over the M2 as the A16 has over the A15 ? That ’s belike an oversimplification , but it ’s the good space to lead off for any discussion of a central processor that ’s still at least a couple of months off .
TSMC’s 3nm process
All the hearsay so far say that the M3 will be a 3 nm chip , manufactured by Apple ’s partner TSMC . That ’s one reason why we might not see an M3 - base product this yr , in fact . TSMC ’s 3 Land of Enchantment summons is still a nascent cognitive process with a comparatively small production capacity , and Apple is using it all just for the A17 in iPhones this fall . Those bigger ( and lower - bulk ) M - series chips want to wait for TSMC ’s manufacturing capacity to increase and that might not chance until 2024 .
But disregardless of timing , this geological fault in manufacture does put a small wrinkle in our performance estimate . The A14 , A15 , A16 , M1 , and M2 were all made on various versions of TSMC ’s 5 nm appendage . The jump to 3 nm is a pretty big spate – it brings far greater density and should allow for either faster clock speeds , lower king , or perhaps a small of both .
The expectant density mean Apple will have the space for more cores , more cache , or other features to make the M3 a big leap from the A16 ’s design than the M2 was to the A15 .
M3: CPU performance estimates
Let ’s start by taking a look at the evolution of unmarried - core CPU performance for recent A - series chip . We ’re looking at a increase of about 9 - 10 % in Geekbench 6.1 , twelvemonth over twelvemonth .
Foundry
Improvements in multi - core performance have been higher , in the 15 - 20 per centum range each class , even though all three buffalo chip have the same core count : two high-pitched - carrying into action core and four high - efficiency core .
How is this move to translate into M3 performance ? We start with the M2 scores and then add the same performance conflict we envision between the A15 ( upon which the M2 is base ) and A16 ( upon which the M3 may be ground ) . And this is what we find …
A undivided - essence musical score of around 2900 or so would put the M3 in the same conference as Intel ’s thirteenth - generation ( Raptor Lake ) central processor . The fastest Intel and AMD background CPUs today mark around 3,000 or just a bit less , and the highest - performing buffalo chip for slight - and - light laptops are in the 2,500 - 2,600 range . Intel and AMD have made big improvements since the M1 take aback the processor diligence with its performance and efficiency , and Apple will need to deliver a astonishingly giving public presentation leap in exclusive - core public presentation with M3 to stay put ahead .
A multi - core Geekbench score of over 11,000 for the M3 accept that Apple delivers the same multi - core performance boost it did with the A16 , but doesnotassume an increase in core count over the M2 . The same goes for the M3 Max score of over 16,000 . While those are undecomposed score , the best screen background mainframe from Intel and AMD currently score around 20,000 .
The 16,000 score is in stock with the best Intel / AMD laptop computer buffalo chip today , and 11,000 is only militant if you wait at very might - constrained competitors .
However , Bloomberg ’s Mark Gurman reportedthat a translation of the M3 Pro in testing now has 12 cores : six high - performance and six high - efficiency . Every M2 has four efficiency scores , so it could be that Apple is adding a span of efficiency cores to at least the M3 Pro / Max and perhaps the entire M3 agate line . Should that happen , you might see multi - core performance go up by 1,000 points or so .
By the meter M3 product gain the market , Intel and AMD are look to have new - genesis processors useable that will raise the performance and efficiency bar even further . When it come to multi - core performance , the M3 generation would have to make larger - than - look improvements in ordination to outpace the advances Intel and AMD are making .
M3 and M3 Max GPU performance estimates
When foretell GPU performance , things vary a little . Apple added GPU cores to the M2 and looks potential to do the same with the M3 generation . The M2 make the max GPU cores from eight to 10 , the M2 Pro from 16 to 19 , and the M2 Max from 32 to 38 . ( The M2 Ultra is two M2 Max french fries stitched together . )
That Mark Gurman leak said that base - mannikin M3 Pro had 18 GPU cores , an gain of two core over the groundwork - model M2 Pro ; the fully fit M3 Pro may therefore have 20 - 24 . From that we can infer the M3 would have 10 - 12 , and the M3 Max 40 - 48 GPU cores .
Here ’s what GPU Compute carrying into action in Geekbench 6 reckon like on Apple ’s A - series flake .
What ’s that lead to do to GPU functioning ? Our estimate takes the leap from A15 to A16 and applies that to the M series .
For GPU compute operation using the Metal API , that entail score approaching 45,000 for the M3 and over 155,000 for the M3 Max . These are estimable figures , butfarbehind the unspoiled graphics cards over on the personal computer side , which score between 250 - 300k on the OpenCL version of this bench mark . The full Radeon cards used with Intel - based Macs grievance over 200,000 . you could lend about 10 pct to those scores if Apple increases the core count as gestate .
Of of course , none of those will fit in a laptop computer . tweet as much GPU performance as you may into a individual system - on - chip with shared memory is Apple ’s game now , and that means great power and caloric execution . The fast scotch are going to be dominate by office - hungry desktop part that can consecrate hundreds of watts each to the CPU and GPU , and that ’s now alone the arena of Windows and Linux PCs .
The speed at which a GPU can run computational algorithms is one thing , but what really gets multitude excited is 3D graphic performance for high - end game . graphic carrying out meliorate a lot pass from the A14 to A15 thanks to the accession of a fifth GPU core , but the A16 still managed to squeeze out more than 12 percent more carrying into action while keeping the same Congress of Racial Equality count thanks largely to the habit of fast memory .
The M serial publication keep adding centre and is expected to bring them again . Our estimate shows a respectable encouragement in performance , but again , this does n’t come anywhere close to exponent - thirsty graphic carte du jour . Even if you supply 10 percent to our idea for additional cores .
A score of 28,000 for the M3 Max frame it in the same ballpark as a GeForce RTX 3070 . Impressive for a big system - on - chip that can go into a laptop computer , but conveniently outclassed by the good background GPUs today by a wide margin .
Apple might yet surprise us
Of course , our carrying out assumption take a feel at the jump from the A15 to A16 and adopt the same leap from M2 to M3 . But if all the rumors are true , the M3 will derail to a raw 3 nm manufacturing process , which can potentially take into account for a big increase in transistor count . The dependable bet is that the M3 will be to the M2 on the dot what the A16 was to the A15 , but I ’m hoping for more .
This could manifest in stack of ways : more cores , more cache , or even fundamental design changes to increase throughput . The rumors of a dyad of special efficiency essence are reasonable , but there would be room for one or two more high - performance cores . Or , the high - performance cores may not trace the architecture of the A16 at all – maybe Apple had metre to comprise the next design , the one likely to appear in the A17 .
The GPU computer architecture is getting a footling old and did n’t really change in the A16 generation . Perhaps M3 will have an entirely new GPU essence design ? Is it too much to hope for electron beam - tracing quickening ?
And of course , AI and machine learning are super important to Apple . An upgraded Neural Engine is not at all out of the question .
I complained that theA16 was essentially an A15 + , a chip with piffling more to show for it than ramped - up clock speeds and faster memory . It has been suppose that the original A16 aim was more divergent from A15 and was intended to be a 3 nm chip . But TSMC ’s appendage was n’t ready in meter to make ten of millions of iPhones last drop , so we got a tuned - up A15 alternatively . But with M3 possibly come up much afterwards , perhaps it will utilize some of the pattern changes and enhancements originally intended for A16 that never made it .
When the M1 was released , the entire CPU industry was prepare back on its dog . By the clip the M3 is released , that will be over three years in the past tense , and both Intel and AMD have had time to make adjustments , pushing carrying into action and efficiency to new height . If Apple is move to continue to strike with its performance and efficiency , the M3 is go to have to deliver more than the same performance leap we found in the M2 . It will need to be more than the A15 - to - A16 transition reassign to laptop and desktop chips . The jump to a 3 New Mexico process affords Apple the opportunity to do a lot more to split up the M3 from its predecessor than it did with the 5 nm A16 , and to protect its reputation , it has to .
