By: Adrian (a.delete@this.acm.org), October 10, 2020 2:18 am
Room: Moderated Discussions
Adrian (a.delete@this.acm.org) on October 10, 2020 1:59 am wrote:
> Mr. Camel (no.thanks.delete@this.gmail.com) on October 9, 2020 6:30 pm wrote:
> > Blue (blue.delete@this.blue.com) on October 8, 2020 9:58 am wrote:
> > > AMD's announce Zen3 with an average of 19% IPC uplift "in desktop applications"
> > > and about a 5-10% clockspeed boost (ST, MT unknown).
> > >
> > > Seems like they're rightfully pushing to up their margin.
> > >
> > > If AMD can keep up this 15-19% gains every generation, when do
> > > the "big on ARM" folks think ARM will effectively catch up?
> >
> > How would an 8 core Zen 3 desktop chip compare to a desktop class 8 core Willow Cove chip?
> >
> > This assumes that a desktop Willow Cove could hit higher frequencies
> > than the mobile version due to more thermal headroom.
> >
> >
>
> The single-thread Cinebench showed Zen 3 @ 4.8 GHz as faster
> by 6% than Willow Cove (Tiger Lake) at the same 4.8 GHz.
>
> So Rocket Lake (it is not known whether its Cypress Cove core is more similar to Willow
> Cove or to its predecessor from Ice Lake, but the differences between these 2 cores are
> small anyway) will have to reach at least 5.3 GHz to slightly exceed a 4.9 GHz Zen 3.
>
>
> For multi-threaded applications an 8-core Rocket Lake will be limited to a lower base frequency than Comet
> Lake (because of doing more work per clock cycle) and it is unlikely that it would match even a 12-core
> Zen 2, and certainly it will be easily beaten by an 8-core Zen 3, due to its higher energy efficiency leading
> to higher clock frequencies when limited by the TDP. The low efficiency of Rocket Lake for power-limited
> MT tasks will be mitigated by 125 W vs. 105 W, but compared to Zen 3 this will not be enough.
>
>
>
>
>
>
Sorry, I might have misinterpreted your question, because the only "Willow Cove" like desktop CPU announced on Intel's roadmaps is the 14-nm Rocket Lake with Cypress Cove.
If your question was about a *10-nm* Intel desktop CPU, i.e. a desktop version of the Tiger Lake H, expected to be launched next year with a TDP up to 65 W, then the answer is different.
Tiger Lake H is expected in laptops, but it might also show up in small desktops, e.g. in a successor of the Intel NUC Pro model that currently uses a Coffee Lake Refresh H 8-core CPU.
Tiger Lake and Zen 3 should have similar energy efficiencies, so they should be close in multi-threaded benchmarks at the same number of cores.
We do not know yet whether an 8-core Tiger Lake will beat an 8-core Zen 3 in MT benchmarks or vice versa, but they should not differ much anyway.
Of course, 12-core and 16-core AMD CPUs will be much faster.
However, despite the higher TDP, I do not believe that a desktop Tiger Lake H may have a much higher turbo than the current 4.8 GHz of Tiger Lake U.
The 4.8 GHz seems to be a limit due to the process, like the 4.9 GHz of the Zen 3 and the active core at that frequency consumes much less than allowed by the thermal headroom.
Until March 2021, Intel might tweak a little their process and their masks, so I expect a top speed of either 4.9 GHz or at most 5.0 GHz for Tiger Lake H, which will have thus frequency parity with Zen 3, but Zen 3 appears to have a better IPC in ST tasks.
> Mr. Camel (no.thanks.delete@this.gmail.com) on October 9, 2020 6:30 pm wrote:
> > Blue (blue.delete@this.blue.com) on October 8, 2020 9:58 am wrote:
> > > AMD's announce Zen3 with an average of 19% IPC uplift "in desktop applications"
> > > and about a 5-10% clockspeed boost (ST, MT unknown).
> > >
> > > Seems like they're rightfully pushing to up their margin.
> > >
> > > If AMD can keep up this 15-19% gains every generation, when do
> > > the "big on ARM" folks think ARM will effectively catch up?
> >
> > How would an 8 core Zen 3 desktop chip compare to a desktop class 8 core Willow Cove chip?
> >
> > This assumes that a desktop Willow Cove could hit higher frequencies
> > than the mobile version due to more thermal headroom.
> >
> >
>
> The single-thread Cinebench showed Zen 3 @ 4.8 GHz as faster
> by 6% than Willow Cove (Tiger Lake) at the same 4.8 GHz.
>
> So Rocket Lake (it is not known whether its Cypress Cove core is more similar to Willow
> Cove or to its predecessor from Ice Lake, but the differences between these 2 cores are
> small anyway) will have to reach at least 5.3 GHz to slightly exceed a 4.9 GHz Zen 3.
>
>
> For multi-threaded applications an 8-core Rocket Lake will be limited to a lower base frequency than Comet
> Lake (because of doing more work per clock cycle) and it is unlikely that it would match even a 12-core
> Zen 2, and certainly it will be easily beaten by an 8-core Zen 3, due to its higher energy efficiency leading
> to higher clock frequencies when limited by the TDP. The low efficiency of Rocket Lake for power-limited
> MT tasks will be mitigated by 125 W vs. 105 W, but compared to Zen 3 this will not be enough.
>
>
>
>
>
>
Sorry, I might have misinterpreted your question, because the only "Willow Cove" like desktop CPU announced on Intel's roadmaps is the 14-nm Rocket Lake with Cypress Cove.
If your question was about a *10-nm* Intel desktop CPU, i.e. a desktop version of the Tiger Lake H, expected to be launched next year with a TDP up to 65 W, then the answer is different.
Tiger Lake H is expected in laptops, but it might also show up in small desktops, e.g. in a successor of the Intel NUC Pro model that currently uses a Coffee Lake Refresh H 8-core CPU.
Tiger Lake and Zen 3 should have similar energy efficiencies, so they should be close in multi-threaded benchmarks at the same number of cores.
We do not know yet whether an 8-core Tiger Lake will beat an 8-core Zen 3 in MT benchmarks or vice versa, but they should not differ much anyway.
Of course, 12-core and 16-core AMD CPUs will be much faster.
However, despite the higher TDP, I do not believe that a desktop Tiger Lake H may have a much higher turbo than the current 4.8 GHz of Tiger Lake U.
The 4.8 GHz seems to be a limit due to the process, like the 4.9 GHz of the Zen 3 and the active core at that frequency consumes much less than allowed by the thermal headroom.
Until March 2021, Intel might tweak a little their process and their masks, so I expect a top speed of either 4.9 GHz or at most 5.0 GHz for Tiger Lake H, which will have thus frequency parity with Zen 3, but Zen 3 appears to have a better IPC in ST tasks.
Topic | Posted By | Date |
---|---|---|
Zen 3 | Blue | 2020/10/08 09:58 AM |
Zen 3 | Rayla | 2020/10/08 10:10 AM |
Zen 3 | Adrian | 2020/10/08 10:13 AM |
Does anyone know whether Zen 3 has AVX-512? (NT) | Foo_ | 2020/10/08 11:54 AM |
Does anyone know whether Zen 3 has AVX-512? | Adrian | 2020/10/08 12:11 PM |
Zen 3 - Number of load/store units | ⚛ | 2020/10/08 10:21 AM |
Zen 3 - Number of load/store units | Rayla | 2020/10/08 10:28 AM |
Zen 3 - Number of load/store units | ⚛ | 2020/10/08 11:22 AM |
Zen 3 - Number of load/store units | Adrian | 2020/10/08 11:53 AM |
Zen 3 - Number of load/store units | Travis Downs | 2020/10/08 09:45 PM |
Zen 3 - CAD benchmark | Per Hesselgren | 2020/10/09 07:29 AM |
Zen 3 - CAD benchmark | Adrian | 2020/10/09 09:27 AM |
Zen 3 - Number of load/store units | itsmydamnation | 2020/10/08 02:38 PM |
Zen 3 - Number of load/store units | Groo | 2020/10/08 02:48 PM |
Zen 3 - Number of load/store units | Wilco | 2020/10/08 03:02 PM |
Zen 3 - Number of load/store units | Dummond D. Slow | 2020/10/08 04:39 PM |
Zen 3 - Number of load/store units | Doug S | 2020/10/09 08:11 AM |
Zen 3 - Number of load/store units | Dummond D. Slow | 2020/10/09 09:43 AM |
Zen 3 - Number of load/store units | Doug S | 2020/10/09 01:43 PM |
N7 and N7P are not load/Store units - please fix the topic in your replies (NT) | Heikki Kultala | 2020/10/10 07:37 AM |
Zen 3 | Jeff S. | 2020/10/08 12:16 PM |
Zen 3 | anon | 2020/10/08 01:57 PM |
Disappointing opening line in paper | Paul A. Clayton | 2020/10/11 06:16 AM |
Thoughts on "Improving the Utilization of µop Caches..." | Paul A. Clayton | 2020/10/14 12:11 PM |
Thoughts on "Improving the Utilization of µop Caches..." | anon | 2020/10/15 11:56 AM |
Thoughts on "Improving the Utilization of µop Caches..." | anon | 2020/10/15 11:57 AM |
Sorry about the mess | anon | 2020/10/15 11:58 AM |
Sorry about the mess | Brett | 2020/10/16 03:22 AM |
Caching dependence info in µop cache | Paul A. Clayton | 2020/10/16 06:20 AM |
Caching dependence info in µop cache | anon | 2020/10/16 12:36 PM |
Caching dependence info in µop cache | Paul A. Clayton | 2020/10/18 01:28 PM |
Zen 3 | juanrga | 2020/10/09 10:12 AM |
Zen 3 | Mr. Camel | 2020/10/09 06:30 PM |
Zen 3 | anon.1 | 2020/10/10 12:44 AM |
Cinebench is terrible benchmark | David Kanter | 2020/10/10 10:36 AM |
Cinebench is terrible benchmark | anon.1 | 2020/10/10 12:06 PM |
Cinebench is terrible benchmark | hobold | 2020/10/10 12:33 PM |
Some comments on benchmarks | Paul A. Clayton | 2020/10/14 12:11 PM |
Some comments on benchmarks | Mark Roulo | 2020/10/14 03:21 PM |
Zen 3 | Adrian | 2020/10/10 01:59 AM |
Zen 3 | Adrian | 2020/10/10 02:18 AM |
Zen 3 | majord | 2020/10/15 04:02 AM |
Zen 3 | hobold | 2020/10/10 08:58 AM |
Zen 3 | Maynard Handley | 2020/10/10 10:36 AM |
Zen 3 | hobold | 2020/10/10 12:19 PM |
Zen 3 | anon | 2020/10/11 02:58 AM |
Zen 3 | hobold | 2020/10/11 12:32 PM |
Zen 3 | anon | 2020/10/11 01:07 PM |
Zen 3 | hobold | 2020/10/11 02:22 PM |
Zen 3 | anon | 2020/10/10 11:51 AM |
Zen 3 | Michael S | 2020/10/11 01:16 AM |
Zen 3 | hobold | 2020/10/11 02:13 AM |
Zen 3 | Michael S | 2020/10/11 02:18 AM |
Zen 3 | anon.1 | 2020/10/11 12:17 PM |
Zen 3 | David Hess | 2020/10/12 06:43 AM |
more power? (NT) | anonymous2 | 2020/10/12 01:26 PM |
I think he's comparing 65W 3700X vs 105W 5800X (NT) | John H | 2020/10/12 04:33 PM |
?! Those are apples and oranges! (NT) | anon | 2020/10/12 04:49 PM |