Info TOP 20 of the World's Most Powerful CPU Cores - IPC/PPC comparison

Page 11 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Richie Rich

Senior member
Jul 28, 2019
470
229
76
Added cores:
  • A53 - little core used in some low-end smartphones in 8-core config (Snapdragon 450)
  • A55 - used as little core in every modern Android SoC
  • A72 - "high" end Cortex core used in Snapdragon 625 or Raspberry Pi 4
  • A73 - "high" end Cortex core
  • A75 - "high" end Cortex core
  • Bulldozer - infamous AMD core
Geekbench 5.1 PPC chart 6/23/2020:

Pos
Man
CPU
Core
Year
ISA
GB5 Score
GHz
PPC (score/GHz)
Relative to 9900K
Relative to Zen3
1​
Nuvia​
(Est.)​
Phoenix (Est.)​
2021​
ARMv9.0​
2001​
3.00​
667.00​
241.0%​
194.1%​
2​
Apple​
A15 (est.)​
(Est.)​
2021​
ARMv9.0​
1925​
3.00​
641.70​
231.8%​
186.8%​
3​
Apple​
A14 (est.)​
Firestorm​
2020​
ARMv8.6​
1562​
2.80​
558.00​
201.6%​
162.4%​
4​
Apple​
A13​
Lightning​
2019​
ARMv8.4​
1332​
2.65​
502.64​
181.6%​
146.3%​
5​
Apple​
A12​
Vortex​
2018​
ARMv8.3​
1116​
2.53​
441.11​
159.4%​
128.4%​
6​
ARM Cortex​
V1 (est.)​
Zeus​
2020​
ARMv8.6​
1287​
3.00​
428.87​
154.9%​
124.8%​
7​
ARM Cortex​
N2 (est.)​
Perseus​
2021​
ARMv9.0​
1201​
3.00​
400.28​
144.6%​
116.5%​
8​
Apple​
A11​
Monsoon​
2017​
ARMv8.2​
933​
2.39​
390.38​
141.0%​
113.6%​
9​
Intel​
(Est.)​
Golden Cove (Est.)​
2021​
x86-64​
1780​
4.60​
386.98​
139.8%​
112.6%​
10​
ARM Cortex​
X1​
Hera​
2020​
ARMv8.2​
1115​
3.00​
371.69​
134.3%​
108.2%​
11
AMD
5900X (Est.)
Zen 3 (Est.)
2020
x86-64
1683
4.90
343.57
124.1%
100.0%
12​
Apple​
A10​
Hurricane​
2016​
ARMv8.1​
770​
2.34​
329.06​
118.9%​
95.8%​
13​
Intel​
1065G7​
Icelake​
2019​
x86-64​
1252​
3.90​
321.03​
116.0%​
93.4%​
14​
ARM Cortex​
A78​
Hercules​
2020​
ARMv8.2​
918​
3.00​
305.93​
110.5%​
89.0%​
15​
Apple​
A9​
Twister​
2015​
ARMv8.0​
564​
1.85​
304.86​
110.1%​
88.7%​
16
AMD
3950X
Zen 2
2019
x86-64
1317
4.60
286.30
103.4%
83.3%
17​
ARM Cortex​
A77​
Deimos​
2019​
ARMv8.2​
812​
2.84​
285.92​
103.3%​
83.2%​
18​
Intel​
9900K​
Coffee LakeR​
2018​
x86-64​
1384​
5.00​
276.80​
100.0%​
80.6%​
19​
Intel​
10900K​
Comet Lake​
2020​
x86-64​
1465​
5.30​
276.42​
99.9%​
80.5%​
20​
Intel​
6700K​
Skylake​
2015​
x86-64​
1032​
4.00​
258.00​
93.2%​
75.1%​
21​
ARM Cortex​
A76​
Enyo​
2018​
ARMv8.2​
720​
2.84​
253.52​
91.6%​
73.8%​
22​
Intel​
4770K​
Haswell​
2013​
x86-64​
966​
3.90​
247.69​
89.5%​
72.1%​
23​
AMD​
1800X​
Zen 1​
2017​
x86-64​
935​
3.90​
239.74​
86.6%​
69.8%​
24​
Apple​
A13​
Thunder​
2019​
ARMv8.4​
400​
1.73​
231.25​
83.5%​
67.3%​
25​
Apple​
A8​
Typhoon​
2014​
ARMv8.0​
323​
1.40​
230.71​
83.4%​
67.2%​
26​
Intel​
3770K​
Ivy Bridge​
2012​
x86-64​
764​
3.50​
218.29​
78.9%​
63.5%​
27​
Apple​
A7​
Cyclone​
2013​
ARMv8.0​
270​
1.30​
207.69​
75.0%​
60.5%​
28​
Intel​
2700K​
Sandy Bridge​
2011​
x86-64​
723​
3.50​
206.57​
74.6%​
60.1%​
29​
ARM Cortex​
A75​
Prometheus​
2017​
ARMv8.2​
505​
2.80​
180.36​
65.2%​
52.5%​
30​
ARM Cortex​
A73​
Artemis​
2016​
ARMv8.0​
380​
2.45​
155.10​
56.0%​
45.1%​
31​
ARM Cortex​
A72​
Maya​
2015​
ARMv8.0​
259​
1.80​
143.89​
52.0%​
41.9%​
32​
Intel​
E6600​
Core2​
2006​
x86-64​
338​
2.40​
140.83​
50.9%​
41.0%​
33​
AMD​
FX-8350​
BD​
2011​
x86-64​
566​
4.20​
134.76​
48.7%​
39.2%​
34​
AMD​
Phenom 965 BE​
K10.5​
2006​
x86-64​
496​
3.70​
134.05​
48.4%​
39.0%​
35​
ARM Cortex​
A57 (est.)​
Atlas​
0​
ARMv8.0​
222​
1.80​
123.33​
44.6%​
35.9%​
36​
ARM Cortex​
A15 (est.)​
Eagle​
0​
ARMv7 32-bit​
188​
1.80​
104.65​
37.8%​
30.5%​
37​
AMD​
Athlon 64 X2 3800+​
K8​
2005​
x86-64​
207​
2.00​
103.50​
37.4%​
30.1%​
38​
ARM Cortex​
A17 (est.)​
0​
ARMv7 32-bit​
182​
1.80​
100.91​
36.5%​
29.4%​
39​
ARM Cortex​
A55​
Ananke​
2017​
ARMv8.2​
155​
1.60​
96.88​
35.0%​
28.2%​
40​
ARM Cortex​
A53​
Apollo​
2012​
ARMv8.0​
148​
1.80​
82.22​
29.7%​
23.9%​
41​
Intel​
Pentium D​
P4​
2005​
x86-64​
228​
3.40​
67.06​
24.2%​
19.5%​
42​
ARM Cortex​
A7 (est.)​
Kingfisher​
0​
ARMv7 32-bit​
101​
1.80​
56.06​
20.3%​
16.3%​

GB5-PPC-evolution.png

GB5-STperf-evolution.png

TOP10PPC_CPU_frequency_evolution_graph.png



TOP 10 - Performance Per Area comparison at ISO-clock (PPA/GHz)

Copied from locked thread. They try to avoid people to see this comparison how x86 is so bad.[/B]

Pos
Man
CPU
Core
Core Area mm2
Year
ISA
SPEC PPA/Ghz
Relative
1​
ARM Cortex​
A78​
Hercules​
1.33​
2020​
ARMv8​
9.41​
100.0%​
2​
ARM Cortex​
A77​
Deimos​
1.40​
2019​
ARMv8​
8.36​
88.8%​
3​
ARM Cortex​
A76​
Enyo​
1.20​
2018​
ARMv8​
7.82​
83.1%​
4​
ARM Cortex​
X1​
Hera​
2.11​
2020​
ARMv8​
7.24​
76.9%​
5​
Apple​
A12​
Vortex​
4.03​
2018​
ARMv8​
4.44​
47.2%​
6​
Apple​
A13​
Lightning​
4.53​
2019​
ARMv8​
4.40​
46.7%​
7​
AMD​
3950X​
Zen 2​
3.60​
2019​
x86-64​
3.02​
32.1%​



It's impressive how fast are evolving the generic Cortex cores:
  • A72 (2015) which can be found in most SBC has 1/3 of IPC of new Cortex X1 - They trippled IPC in just 5 years.
  • A73 and A75 (2017) which is inside majority of Android smart phones today has 1/2 IPC of new Cortex X1 - They doubled IPC in 3 years.

Comparison how x86 vs. Cortex cores:
  • A75 (2017) compared to Zen1 (2017) is loosing massive -34% PPC to x86. As expected.
  • A77 (2019) compared to Zen2 (2018) closed the gap and is equal in PPC. Surprising. Cortex cores caught x86 cores.
  • X1 (2020) is another +30% IPC over A77. Zen3 need to bring 30% IPC jump to stay on par with X1.

Comparison to Apple cores:
  • AMD's Zen2 core is slower than Apple's A9 from 2015.... so AMD is 4 years behind Apple
  • Intel's Sunny Cove core in Ice Lake is slower than Apple's A10 from 2016... so Intel is 3 years behind Apple
  • Cortex A77 core is slower than Apple's A9 from 2015.... but
  • New Cortex X1 core is slower than Apple's A11 from 2017 so ARM LLC is 3 years behind Apple and getting closer



GeekBench5.1 comparison from 6/22/2020:
  • added Cortex X1 and A78 performance projections from Andrei here
  • 2020 awaiting new Apple A14 Firestorm core and Zen3 core
Updated:



EDIT:
Please note to stop endless discussion about PPC frequency scaling: To have fair and clean comparison I will use only the top (high clocked) version from each core as representation for top performance.
 
Last edited:
  • Like
Reactions: chechito

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
If you build CPUs designed for HPC (as used to be done a few decades ago before hundreds then thousands and beyond of off the shelf RISC or x86 CPUs were done - i.e. Seymour Cray's stuff at CDC and then Cray) by including very wide SIMD instructions and and some fast local memory then they are going to be better than GPUs because they are designed for computation, not graphics with a side of "oh hey we'll support HPC too because we found out we can charge way more for HPC cards than GPUs".

That seems to be the direction Fujitsu is taking with A64FX, and others are welcome to try, but there are reasons why dGPUs have taken over in HPC. And it's not that the programming model for external compute cards is particularly fun to use.

Mostly it has to do with the availability of resources to commit to design and implementation, as well as the applicability of designs to multiple markets. It's easier to "cobble together" a supercomputer out of server CPUs based on commodity desktop core designs and dGPUs based at least loosely on commodity desktop dGPU designs than it is custom stuff like A64FX because A64FX has exactly one market: HPC. Design costs on cutting-edge nodes are pretty high. I can sell (for example) Zen3 cores all across server, workstation, desktop, and mobile markets, but I can't necessarily sell designs derivative of A64FX in the same way. I can basically say the same thing about NV's designs (some of the work they put into their high-end compute cards also goes into their consumer dGPUs serving many markets).
 

Hitman928

Diamond Member
Apr 15, 2012
5,177
7,628
136
64 core Rome costs $7500 because that's what AMD is able to get people to pay for it. Surely you aren't so naive you think the prices that Intel and AMD charge for their top end CPUs has anything whatsoever to do with the price to make them?

Teardowns/BOMs of iPhones have consistently pegged the cost of Apple's SoCs in the $30 range, and they are generally around 100 mm^2. Let's say the firms that do these things for a living don't know what they are doing, and the price is actually double that, so $60. What's the total silicon area of a 64C Rome? Seems pretty difficult to make a compelling case the foundry cost for Rome is more than $500 and even that may be high.

Now that figure doesn't include amortization of design cost, all the testing/validation servers require and other fixed costs that are divided up across all units (including SKUs with fewer cores) this is just how much you have to figure they are paying TSMC for their part.

As Markfw says, the math for a monolithic chip is different because the bigger a chip the worse the yield. They can get around that by having redundancy (i.e. extra cores) but people here are claiming that Graviton isn't doing that.

You do realize that I was referring to Richie Rich's claim that Rome costs $7500 and I was criticizing it, that claim didn't come from me.

Having redundancy can help yields but it's not a complete panacea, you'll still get defective chips because of the location of the defect(s). Even if you take defects out of the equation, my original comments on the subject (if you go back and read) still stand. I never once tried to peg an absolute price or even come to a firm conclusion, I specifically avoided doing so because of all the unknowns. I was just addressing the ridiculousness of Richie Rich's original claim and explaining why Mark's claim was certainly plausible.
 

MrTeal

Diamond Member
Dec 7, 2003
3,554
1,658
136
That's fine, I just wanted to make sure because the original claim of $500 came from Richie Rich saying that Graviton2 only cost $500 when 64 core Rome cost $7500. I'm not sure what the wafer cost of TSMC's 7 nm process is so I'm not comfortable putting an exact dollar amount on any of them.
We can set some bounds on it though. The 5500 XT 4GB launched at $170 for a fully enabled 158mm² 7nm Navi 14 card. That's helpfully pretty much exactly double the size of a Zen chiplet. As an absolute maximum, a Zen 2 chiplet shouldn't cost more than $85. Given that $170 includes 4GB GDDR5, the PCB and all its components and assembly, all boxing, shipping, and various levels of markup, even $85 is probably 2x too expensive.
If Zen 2 chiplet costs $50, then there's $400 worth of chiplets in a 64 core Rome processor. The 416mm² IO die on 14nm. AMD hasn't produced much on 14nm lately to set an upper bound on cost, but Nvidia launched the 2060 a year ago with 6GB GDDR6 for $350, and that's a 454mm² die on 12nm. Especially given nvidias margins, that die itself is likely well under $200, but say $200 for the IO die as a reasonable upper.

That would give a hard upper limit of $1030 (8*85+350) and a more reasonable upper limit of $600 for production costs on a 64 core Rome. I'd guess it's decently lower than that. These are all super hand wavey numbers, but this is the thread for that. Production costs on Rome and Gravitron 2 likely aren't too out of line.
 
  • Like
Reactions: Tlh97 and name99

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,478
14,434
136
We can set some bounds on it though. The 5500 XT 4GB launched at $170 for a fully enabled 158mm² 7nm Navi 14 card. That's helpfully pretty much exactly double the size of a Zen chiplet. As an absolute maximum, a Zen 2 chiplet shouldn't cost more than $85. Given that $170 includes 4GB GDDR5, the PCB and all its components and assembly, all boxing, shipping, and various levels of markup, even $85 is probably 2x too expensive.
If Zen 2 chiplet costs $50, then there's $400 worth of chiplets in a 64 core Rome processor. The 416mm² IO die on 14nm. AMD hasn't produced much on 14nm lately to set an upper bound on cost, but Nvidia launched the 2060 a year ago with 6GB GDDR6 for $350, and that's a 454mm² die on 12nm. Especially given nvidias margins, that die itself is likely well under $200, but say $200 for the IO die as a reasonable upper.

That would give a hard upper limit of $1030 (8*85+350) and a more reasonable upper limit of $600 for production costs on a 64 core Rome. I'd guess it's decently lower than that. These are all super hand wavey numbers, but this is the thread for that. Production costs on Rome and Gravitron 2 likely aren't too out of line.
Since Graviton2 is a monolitic die, I still say its more than Rome, like possibly twice in MFG cost.

That said, what set me off was saying that Graviton was $500 vs Rome at $7500, cost vs retail. (or there about)
 

Vixis Rei

Junior Member
Jul 4, 2020
13
16
41
We can set some bounds on it though. The 5500 XT 4GB launched at $170 for a fully enabled 158mm² 7nm Navi 14 card. That's helpfully pretty much exactly double the size of a Zen chiplet. As an absolute maximum, a Zen 2 chiplet shouldn't cost more than $85. Given that $170 includes 4GB GDDR5, the PCB and all its components and assembly, all boxing, shipping, and various levels of markup, even $85 is probably 2x too expensive.
If Zen 2 chiplet costs $50, then there's $400 worth of chiplets in a 64 core Rome processor. The 416mm² IO die on 14nm. AMD hasn't produced much on 14nm lately to set an upper bound on cost, but Nvidia launched the 2060 a year ago with 6GB GDDR6 for $350, and that's a 454mm² die on 12nm. Especially given nvidias margins, that die itself is likely well under $200, but say $200 for the IO die as a reasonable upper.

That would give a hard upper limit of $1030 (8*85+350) and a more reasonable upper limit of $600 for production costs on a 64 core Rome. I'd guess it's decently lower than that. These are all super hand wavey numbers, but this is the thread for that. Production costs on Rome and Gravitron 2 likely aren't too out of line.

Just for speculations sake.
On many other places i've heard that a Zen2 chiplet either costs from about $9 to $18. With people citing the lower amount to the recent good yields and the higher amount to the worser yields around release. Which makes sense. The dies are small and they're getting hundreds of yields per wafer

I wouldn't be surprised if the I/O die is about the same price range as well

I'm expecting them to make massive profits off of each CPU sold. Explains why they've been paying off all their massive debts in record time

Since Graviton2 is a monolitic die, I still say its more than Rome, like possibly twice in MFG cost.

That said, what set me off was saying that Graviton was $500 vs Rome at $7500, cost vs retail. (or there about)

Agreed, because if Graviton2 was only $500 how would they even make any kind of profit from it.

That was the very post that made me finally sign up. And because a friend of mines was reading this thread and thought she could do large amounts of render work and video editing on a A13 because of the fancy chart on the first page and the topic title. Gave me a good laugh.
 
Last edited:
  • Like
Reactions: Tlh97

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Agreed, because if Graviton2 was only $500 how would they even make any kind of profit from it.

Remember that Amazon doesn't sell those things. They're being used for internal stuff and/or AWS. Likely no one will have Graviton2 outside of Amazon unless/until they start dumping old hardware on eBay through liquidators when the platform is retired.
 

Doug S

Platinum Member
Feb 8, 2020
2,201
3,405
136
Remember that Amazon doesn't sell those things. They're being used for internal stuff and/or AWS. Likely no one will have Graviton2 outside of Amazon unless/until they start dumping old hardware on eBay through liquidators when the platform is retired.

Yep, the only determinant of whether Graviton2 "makes a profit" is whether switching from buying CPUs from Intel/AMD versus designing your own and having them fabbed saves money. Amazon doesn't care whether they cost $50 or $5000 so long as the whole thing costs them less than buying x86 CPUs did.
 

name99

Senior member
Sep 11, 2010
404
303
136
Yep, the only determinant of whether Graviton2 "makes a profit" is whether switching from buying CPUs from Intel/AMD versus designing your own and having them fabbed saves money. Amazon doesn't care whether they cost $50 or $5000 so long as the whole thing costs them less than buying x86 CPUs did.

All true, but (just like with Apple) the bean counters on these threads continue to miss the big picture.
Amazon, like Apple, are in this for the long haul. They have visions of what they want to do with computing next year, in five years, in fifteen years. And if they control the hardware, they can make that vision come true. Sticking with x86 they are limited to Intel or AMD's timetable, and to their agenda.

Even if Amazon are restricted to using ARM's cores, with ARM everything can be negotiated. They can buy ARM core designs and modify them. They can suggest mods to the ARM ISA (or some other aspect) and if it's a good idea ARM will probably listen (every year a new ARM v8 addendum comes out with various improvements). They can add other stuff to the SoC.
If Amazon thinks the time is right to jump onto Gen-Z, or if they want to add their own TPU onto the SoC, or if they want to put a GPU on their and start processing Ring video, they can do all that, and they can do it on their timetable.

Same with Apple. Cheaper SoCs is nice, lower power/faster performance is nice. But the part that REALLY matters is that they now control the Mac SoC and can improve it, and modify it, at their pace, not at Intel's pace.

You can't capture this in terms like profit; no-one knows what exactly this control will enable in five, let alone fifteen, years, even less so what that means in terms of revenues.
Could you know what the Apple Watch revenues today would be before it was released? Any idea what they will be in 2030 (along with whatever other future products are enabled by Apple's bothering to invest in making smaller core alongside their large iPhone cores)?

But that's the difference between engineering-driven companies like Apple and Amazon and finance-driven companies like Intel and Qualcomm. Finance-driven companies see only certainty, only known markets and known patterns; engineering-driven companies see possibility.
If you don't see possibility, only the certainty of more of today, then of course you won't understand why Apple and Amazon are doing what they are doing.
 
  • Like
Reactions: Tlh97 and teejee

Richie Rich

Senior member
Jul 28, 2019
470
229
76
TOP42 IPC chart

New CPUs added:

Nuvia Phoenix:
  • Released official performance range in GB5 (about 2000 pts @4.5W) here
  • process node N5P TSMC -> estimated frequency 3.0 GHz
  • IPC about 2.41x Intel 9900K
  • uarch similar to Apple A15 (same chief architect Gerard Williams III.)
  • probably 8xALUs + 4xBr (12-wide integer)
  • ISA ARMv9 + SVE2 vectors (range 128 up to 2048-bit)

Apple A14 Firestorm:
  • Released official performance (+40% above A12)
  • process node N5 TSMC -> estimated frequency 2.8 GHz
  • IPC about +11% above A13 Lightning core
  • uarch still 6xALU+3xBr (9-wide integer)... last 6xALU core
  • LPDDR5

Apple A15:
  • probably completely new uarch, first 8xALU core
  • 4xALU core line up (A7, A8, A9, A10)
  • 6xALU core line up (A11, A12, A13, A14)
  • process node N5P
  • IPC similar to Nuvia Phoenix
  • also ARMv9.0 + SVE2 vectors (bfloat16 and matrix multiplication is mandatory for ARMv8.6 + SVE)
  • no Ice Lake, Tiger Lake supports bfloat16 (only Intel Cooper lake as part of AVX512)

ARM Neoverse V1 core (Zeus)
  • based on Cortex X1 core Hera (X1 has +30% IPC over A77 and Zen2)
  • +50% IPC over N1 core (similar IPC to A77) means +50% IPC over Zen2
  • +25% IPC over Zen 3 (estimated +20% IPC over Zen 2)
  • +11% IPC over Golden Cove (estimated +40% IPC over 9900K)
  • SVE SIMD 2x256-bit
  • support Bfloat16 ..... 2x performance for ML
  • FPU IPC +82% is quite brutal (this means integer IPC similar to X1-ish +30%, the +50% IPC comes from FPU side)
  • up to 192-cores
  • chiplet design!!! (AMD will have no advantage in 2021)
  • HBM2E + DDR5 support!!! (first CPU with HBM2E support?)

Intel Golden Cove:
  • rumor IPC +40% over 9900K (rough estimation)
  • added for estimation what we can expect from Alder Lake next year incompare to Nuvia, ARM V1, Apple A15, and Zen 3

AMD Zen 3:
  • only +20% IPC integer (slower to Golden Cove, ARM X1 and V1 core, Apple A14, Nuvia)
  • only +50% IPC in FPU (slower than ARM X1 and V1, Apple, Nuvia)
  • only DDR4
  • the weakest new uarch in terms of IPC for 2021, that might be big problem for AMD
  • no matrix multiplication support for HPC (Intel has AMX for Sapphire Rapids 2021, ARM has SVE since 2019)


Pos
Man
CPU
Core
Year
ISA
GB5 Score
GHz
PPC (score/GHz)
Relative to 9900K
Relative to Zen3
1​
Nuvia​
(Est.)​
Phoenix (Est.)​
2021​
ARMv9.0​
2001​
3.00​
667.00​
241.0%​
194.1%​
2​
Apple​
A15 (Est.)​
(Est.)​
2021​
ARMv9.0​
1925​
3.00​
641.70​
231.8%​
186.8%​
3​
Apple​
A14​
Firestorm​
2020​
ARMv8.6​
1562​
2.80​
558.00​
201.6%​
162.4%​
4​
Apple​
A13​
Lightning​
2019​
ARMv8.4​
1332​
2.65​
502.64​
181.6%​
146.3%​
5​
Apple​
A12​
Vortex​
2018​
ARMv8.3​
1116​
2.53​
441.11​
159.4%​
128.4%​
6​
ARM Cortex​
V1​
Zeus​
2020​
ARMv8.6​
1287​
3.00​
428.87​
154.9%​
124.8%​
7​
ARM Cortex​
N2​
Perseus​
2021​
ARMv9.0​
1201​
3.00​
400.28​
144.6%​
116.5%​
8​
Apple​
A11​
Monsoon​
2017​
ARMv8.2​
933​
2.39​
390.38​
141.0%​
113.6%​
9​
Intel​
(Est.)​
Golden Cove (Est.)​
2021​
x86-64​
1780​
4.60​
386.98​
139.8%​
112.6%​
10​
ARM Cortex​
X1​
Hera​
2020​
ARMv8.2​
1115​
3.00​
371.69​
134.3%​
108.2%​
11
AMD
5900X (Est.)
Zen 3 (Est.)
2020
x86-64
1683
4.90
343.57
124.1%
100.0%
12​
Apple​
A10​
Hurricane​
2016​
ARMv8.1​
770​
2.34​
329.06​
118.9%​
95.8%​
13​
Intel​
1065G7​
Icelake​
2019​
x86-64​
1252​
3.90​
321.03​
116.0%​
93.4%​
14​
ARM Cortex​
A78​
Hercules​
2020​
ARMv8.2​
918​
3.00​
305.93​
110.5%​
89.0%​
15​
Apple​
A9​
Twister​
2015​
ARMv8.0​
564​
1.85​
304.86​
110.1%​
88.7%​
16
AMD
3950X
Zen 2
2019
x86-64
1317
4.60
286.30
103.4%
83.3%
17​
ARM Cortex​
A77​
Deimos​
2019​
ARMv8.2​
812​
2.84​
285.92​
103.3%​
83.2%​
18​
Intel​
9900K​
Coffee LakeR​
2018​
x86-64​
1384​
5.00​
276.80​
100.0%​
80.6%​
19​
Intel​
10900K​
Comet Lake​
2020​
x86-64​
1465​
5.30​
276.42​
99.9%​
80.5%​
20​
Intel​
6700K​
Skylake​
2015​
x86-64​
1032​
4.00​
258.00​
93.2%​
75.1%​
21​
ARM Cortex​
A76​
Enyo​
2018​
ARMv8.2​
720​
2.84​
253.52​
91.6%​
73.8%​
22​
Intel​
4770K​
Haswell​
2013​
x86-64​
966​
3.90​
247.69​
89.5%​
72.1%​
23​
AMD​
1800X​
Zen 1​
2017​
x86-64​
935​
3.90​
239.74​
86.6%​
69.8%​
24​
Apple​
A13​
Thunder​
2019​
ARMv8.4​
400​
1.73​
231.25​
83.5%​
67.3%​
25​
Apple​
A8​
Typhoon​
2014​
ARMv8.0​
323​
1.40​
230.71​
83.4%​
67.2%​
26​
Intel​
3770K​
Ivy Bridge​
2012​
x86-64​
764​
3.50​
218.29​
78.9%​
63.5%​
27​
Apple​
A7​
Cyclone​
2013​
ARMv8.0​
270​
1.30​
207.69​
75.0%​
60.5%​
28​
Intel​
2700K​
Sandy Bridge​
2011​
x86-64​
723​
3.50​
206.57​
74.6%​
60.1%​
29​
ARM Cortex​
A75​
Prometheus​
2017​
ARMv8.2​
505​
2.80​
180.36​
65.2%​
52.5%​
30​
ARM Cortex​
A73​
Artemis​
2016​
ARMv8.0​
380​
2.45​
155.10​
56.0%​
45.1%​
31​
ARM Cortex​
A72​
Maya​
2015​
ARMv8.0​
259​
1.80​
143.89​
52.0%​
41.9%​
32​
Intel​
E6600​
Core2​
2006​
x86-64​
338​
2.40​
140.83​
50.9%​
41.0%​
33​
AMD​
FX-8350​
BD​
2011​
x86-64​
566​
4.20​
134.76​
48.7%​
39.2%​
34​
AMD​
Phenom 965 BE​
K10.5​
2006​
x86-64​
496​
3.70​
134.05​
48.4%​
39.0%​
35​
ARM Cortex​
A57​
Atlas​
0​
ARMv8.0​
222​
1.80​
123.33​
44.6%​
35.9%​
36​
ARM Cortex​
A15​
Eagle​
0​
ARMv7 32-bit​
188​
1.80​
104.65​
37.8%​
30.5%​
37​
AMD​
Athlon 64 X2 3800+​
K8​
2005​
x86-64​
207​
2.00​
103.50​
37.4%​
30.1%​
38​
ARM Cortex​
A17​
0​
ARMv7 32-bit​
182​
1.80​
100.91​
36.5%​
29.4%​
39​
ARM Cortex​
A55​
Ananke​
2017​
ARMv8.2​
155​
1.60​
96.88​
35.0%​
28.2%​
40​
ARM Cortex​
A53​
Apollo​
2012​
ARMv8.0​
148​
1.80​
82.22​
29.7%​
23.9%​
41​
Intel​
Pentium D​
P4​
2005​
x86-64​
228​
3.40​
67.06​
24.2%​
19.5%​
42​
ARM Cortex​
A7​
Kingfisher​
0​
ARMv7 32-bit​
101​
1.80​
56.06​
20.3%​
16.3%​
 
Last edited:
  • Like
Reactions: bearmoo

Doug S

Platinum Member
Feb 8, 2020
2,201
3,405
136
So now he's estimating Nuvia's performance based on a press release with a graph that shows a WIDE range of possible figures, and claims it will be shipping in 2021. And goes even further, deciding what he thinks Apple will do with the A15 and pulling performance estimates for it out of his ass.

As if this thread wasn't stupid enough.
 

Richie Rich

Senior member
Jul 28, 2019
470
229
76
So now he's estimating Nuvia's performance based on a press release with a graph that shows a WIDE range of possible figures, and claims it will be shipping in 2021. And goes even further, deciding what he thinks Apple will do with the A15 and pulling performance estimates for it out of his ass.

As if this thread wasn't stupid enough.
What performance would you suggest?

This is official Nuvia picture:
- at 4Watts is range 1750 pts up to 2200 pts
- at 4.5 Watts is range 1900 up to 2050 pts
- on N5P TSMC process 3.0 GHz is reasonable

  • The worst case scenario: 1750 pts @ 3 GHz = 1750 / 3 = 583 pts/GHz .... 2.04x higher IPC than AMD Zen2
  • The best case scenario: 2200 pts @ 3 GHz = 2200 / 3 = 733 pts/GHz .... 2.54x higher IPC than AMD Zen2
  • My estimation is 2000 pts @ 3.0 GHz. Roughly in the middle.

However even the worst case shows brutal IPC. Even much higher than current king Apple A13. Which is not surprising as Nuvia is lead by ex-Apple chief architect responsible for Apple 6xALU A11-A14 line up (he left Apple in 2018 so probably responsible for A15 and A16 too).
Nuvia_Phoenix_ST_performance_prediction.png
 

scannall

Golden Member
Jan 1, 2012
1,944
1,638
136
None. Wait until actual hardware goes to reviewers. I have other suggestions, that you will also likely ignore.
Exactly. For instance, I'm sure the upcoming ARM based Macs will perform really well. But how well? Wait for the reviews. If I can't buy it, I'm not interested in guesses and marketing slides.
 

Richie Rich

Senior member
Jul 28, 2019
470
229
76
Exactly. For instance, I'm sure the upcoming ARM based Macs will perform really well. But how well? Wait for the reviews. If I can't buy it, I'm not interested in guesses and marketing slides.
  • CPU A: hundreds of pages discussing every bit of marketing slide about "unified L3 cache" and +15% IPC
  • CPU B: no discussion even marketing slides suggesting IPC to be more than +100% over Zen 2

I guess if AMD or Intel would release same slide with at least +100% IPC there would be 1000 pages about that. Why is that?

Because Nuvia Phoenix and AMD Zen3/4 will compete in servers and workstations next year. Same with ARM V1 core bringing +50% IPC over Zen2, SVE vectors up to 2048-bit width architecture, matrix multiplication, Bfloat16 support (doubling ML performance), chiplet architecture, up to 192-cores, HBM2E + DDR5 memory. This is way better than what Intel Golden Cove or AMD Zen3/4 can offer. Am I the last CPU enthusiast here who cares about raw performance and not about brand?
 
  • Like
Reactions: Lodix and name99

Richie Rich

Senior member
Jul 28, 2019
470
229
76
No you are not
but about the direction of your thought your are worlduniverse size of being wrong
I specifically wrote CPU enthusiast. Not x86 enthusiast. When you want discuss the most powerful CPU you have to go outside x86 garden. For example Fujitsu A64FX with HMB memory and SVE matrix multiplication is number one in supercomputers. Beating Nvidia GPUs.

Which x86 CPU can beat Nvidia Volta GPU? None. ARM A64FX can.
Which x86 CPU has HBM memory? None. ARM A64FX has.
Which x86 CPU has matrix multiplication? None. ARM A64FX has [SVE].
Which x86 SIMD are 2048-bit wide? None. ARM SVE/SVE2 is up to 2048-bit wide.
Which x86 CPU suggests +50% IPC after one year? None. ARM V1 does.
Which x86 CPU suggests +100% IPC over Intel 9900K? None. Nuvia Phoenix does.

Those are facts provided by companies itself. Not by me. I'm just a messenger. Or you suggest Nuvia engineers are wrong?
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Am I the last CPU enthusiast here who cares about raw performance and not about brand?

You are the only one trying to create a bogus IPC list. While other CPUs besides Nuvia products may be "discussed to death", nobody worth their salt is going to try to definitively state IPC/GHz or anything else based on those unreleased CPUs.
 

TheGiant

Senior member
Jun 12, 2017
748
353
106
I specifically wrote CPU enthusiast. Not x86 enthusiast. When you want discuss the most powerful CPU you have to go outside x86 garden. For example Fujitsu A64FX with HMB memory and SVE matrix multiplication is number one in supercomputers. Beating Nvidia GPUs.

Which x86 CPU can beat Nvidia Volta GPU? None. ARM A64FX can.
Which x86 CPU has HBM memory? None. ARM A64FX has.
Which x86 CPU has matrix multiplication? None. ARM A64FX has [SVE].
Which x86 SIMD are 2048-bit wide? None. ARM SVE/SVE2 is up to 2048-bit wide.
Which x86 CPU suggests +50% IPC after one year? None. ARM V1 does.
Which x86 CPU suggests +100% IPC over Intel 9900K? None. Nuvia Phoenix does.

Those are facts provided by companies itself. Not by me. I'm just a messenger. Or you suggest Nuvia engineers are wrong?
maybe I wrote it wrong
you are not the last CPU enthusiast to care about results and interpreted in specific tools
the message was the the results are not on the first place
 
  • Like
Reactions: Tlh97

Thala

Golden Member
Nov 12, 2014
1,355
653
136
You are the only one trying to create a bogus IPC list. While other CPUs besides Nuvia products may be "discussed to death", nobody worth their salt is going to try to definitively state IPC/GHz or anything else based on those unreleased CPUs.

Why is this bogus? Looks pretty accurate to me.
I am not the biggest fan on adding unreleased products to the list - but why not if it is marked as estimation?