Arctic Liquid Freezer II 280 liquid cooler review
Promo: Windows 10 Pro licenses for under 10 USD
Team Group MP33 NVMe 512 GB SSD Review
ADATA SE800 1TB Portable SSD review
Promo: Microsoft Office 2016 for $29.82
Corsair H100i RGB Pro XT liquid cooler review
Corsair K95 RGB PLATINUM XT review
AORUS Radeon RX 5700 XT 8G review
Endgame XM1 gaming mouse review
Guru3D Rig of the Month - December 2019
Eight core Coffee Lake Processor Finally Spotted?
It has been topic of discussion for many months now, ever since the 'Z390 chipset' surfaced on a roadmap and a company kind of leaked that it was intended for an 8-core processor, it became silent though, from the looks of it, it's for real.
Over at Videocardz, they spotted an entry in the Futuremark, and indeed that very much looks like an 8-core processor that is actually listed as Coffee Lake. The processor is listed as a generic, and that means the motherboard firmware microcode has not been updated or recognizing the processor. That base-clock could be misread, or this is a very early engineering sample.
The Z390 chipset has been spotted on several roadmaps and slides, so at one point we do need to understand what it is intended for eh? Anyway, the CPU is listed at 8-cores and 16-threads, the rest of the stuff looks erroneous, including the listed base clock. The testing platform is listed as Intel Corporation CoffeeLake S82 UDIMM RVP, likely an engineering board.
It will not be released in April though, that moment is for H370 and B360 chipsets and a few new Coffee Lake respin procs, this later. Also, there has been a rumor about the Z370 chipset, it would have been incompatible with Skylake and Kaby Lake due to Intel's preparations for an 8-core CPU.
Make of it what you will, but we're certain of one fact, Intel does not like the fact that AMD has 8-core processors available in the mainstream segment, while they do not.
Benchmark show Probable Eight Core Ryzen Zen+ 12nm at 4.35 GHz - 03/07/2018 08:08 AM
With each week that passes we hear a little more about the upcoming Zen+ 12nm Ryzern update. The biggest question, will AMD be able to achieve clock frequencies that are higher? Well, a new set of lea...
Eight Core Ryzen Threadripper 1900X Spotted in India Etail - 08/27/2017 04:06 PM
That third announced Ryzen Threadripper processor called the 1900X is spotted in an India Etail stores.The 1900X is the 8-cores part and will be the 'lowest' SKU available in the Threadripper serie...
Intel Poulson Itanium to get eight cores - 02/22/2011 11:38 AM
Intel's Itanium series will soon get an update with an eight-core part called Poulson (codename). This 32nm CPU will be the world's largest general-purpose processor designed to date, it packs 3.1 bil...
Fox2232
Senior Member
Posts: 9837
Joined: 2012-07-20
Senior Member
Posts: 9837
Joined: 2012-07-20
#5530281 Posted on: 03/21/2018 10:28 PM
Ryzen and Epyc are different platforms, so stop comparing apples and oranges to get your points across. /benchmarks I saw tells the different story, depends on a number of threads, sometimes AMD is better, sometimes Intel, one AMD CCX /4cores/ can access only up to 30GB/s /
8700k and 1800x have pretty much same memory bandwith while 8700k having 3x lower latencies, also i7 8700k's cache manages higher bandwith, so not really. OC 8700k is almost as quick or even faster as an 8 core Ryzen in cpu heavy tasks such as rendering etc. Intel also generally handles and supports higher RAM clocks and lower latencies better.
None of current Intel cpus were developed as quad cores, all cores all native and the ringbus is scalable, the culprit is that at a certain point you start encountering higher latencies due to a higher number of cycles, so Intel is upgrading to the mesh.
If anything, AMD current cpus were designed mainly as quadcores which are interconnected together in order to be scaled, look at their latencies and their gaming performance being heavily affected by RAM clocks/timings. So stop with this projecting towards Intel.
So I really don't get it were are you getting your assumptions from because by your logic, even 8700k should be somehow underpeforming and suffering from "the quad core design" and "missing the quad channel". But the reality is different and 8700k is riping everything apart in games and contests with 8cores when OC-ed at rendering.
Isn't it funny when you look at it, but can't see it. You even wrote it down:
"8700k and 1800x have pretty much same memory bandwith"
What you do not get is Bandwidth scaling. Ryzen AM4 has 20~35% bandwidth scaling when you move from ST to MT workloads. Coffee Lake 7~13% bandwidth Scaling from ST to MT.
Those Higher end platforms with Quad channel have much better bandwidth scaling. So, now Intel brings 8C/16T chip with dual channel which does not scale bandwidth with increased number of cores used.
For those who will not want to have this chip bandwidth starved at certain workloads, it will be very cheap chip. Because cost of the top OC memory kit will make it look that way.
Want to see if it will be memory starved? Ask someone with Coffee 6C/12T to reduce memory clock from 3200MHz to 2400MHz. Why, because 6C/12T is 25% less cores than 8C/16T. And 2400MHz is 25% lower clock than 3200MHz. (But I think you are already starting to get it, right?)
Ryzen and Epyc are different platforms, so stop comparing apples and oranges to get your points across. /benchmarks I saw tells the different story, depends on a number of threads, sometimes AMD is better, sometimes Intel, one AMD CCX /4cores/ can access only up to 30GB/s /
8700k and 1800x have pretty much same memory bandwith while 8700k having 3x lower latencies, also i7 8700k's cache manages higher bandwith, so not really. OC 8700k is almost as quick or even faster as an 8 core Ryzen in cpu heavy tasks such as rendering etc. Intel also generally handles and supports higher RAM clocks and lower latencies better.
None of current Intel cpus were developed as quad cores, all cores all native and the ringbus is scalable, the culprit is that at a certain point you start encountering higher latencies due to a higher number of cycles, so Intel is upgrading to the mesh.
If anything, AMD current cpus were designed mainly as quadcores which are interconnected together in order to be scaled, look at their latencies and their gaming performance being heavily affected by RAM clocks/timings. So stop with this projecting towards Intel.
So I really don't get it were are you getting your assumptions from because by your logic, even 8700k should be somehow underpeforming and suffering from "the quad core design" and "missing the quad channel". But the reality is different and 8700k is riping everything apart in games and contests with 8cores when OC-ed at rendering.
Isn't it funny when you look at it, but can't see it. You even wrote it down:
"8700k and 1800x have pretty much same memory bandwith"
What you do not get is Bandwidth scaling. Ryzen AM4 has 20~35% bandwidth scaling when you move from ST to MT workloads. Coffee Lake 7~13% bandwidth Scaling from ST to MT.
Those Higher end platforms with Quad channel have much better bandwidth scaling. So, now Intel brings 8C/16T chip with dual channel which does not scale bandwidth with increased number of cores used.
For those who will not want to have this chip bandwidth starved at certain workloads, it will be very cheap chip. Because cost of the top OC memory kit will make it look that way.
Want to see if it will be memory starved? Ask someone with Coffee 6C/12T to reduce memory clock from 3200MHz to 2400MHz. Why, because 6C/12T is 25% less cores than 8C/16T. And 2400MHz is 25% lower clock than 3200MHz. (But I think you are already starting to get it, right?)
nz3777
Senior Member
Posts: 2382
Joined: 2014-01-21
Senior Member
Posts: 2382
Joined: 2014-01-21
#5530840 Posted on: 03/23/2018 01:45 PM
Yeah but Intel use a Special glue not just your average Amd el-cheepo one. What a bunch of Haters these guys are I swear to god.
Yeah but Intel use a Special glue not just your average Amd el-cheepo one. What a bunch of Haters these guys are I swear to god.
Koniakki
Senior Member
Posts: 2843
Joined: 2009-09-15
Senior Member
Posts: 2843
Joined: 2009-09-15
#5531579 Posted on: 03/26/2018 05:46 AM
......
Want to see if it will be memory starved? Ask someone with Coffee 6C/12T to reduce memory clock from 3200MHz to 2400MHz. Why, because 6C/12T is 25% less cores than 8C/16T. And 2400MHz is 25% lower clock than 3200MHz. (But I think you are already starting to get it, right?)
I can help with that. Any specific tests to run?
......
Want to see if it will be memory starved? Ask someone with Coffee 6C/12T to reduce memory clock from 3200MHz to 2400MHz. Why, because 6C/12T is 25% less cores than 8C/16T. And 2400MHz is 25% lower clock than 3200MHz. (But I think you are already starting to get it, right?)
I can help with that. Any specific tests to run?
Fox2232
Senior Member
Posts: 9837
Joined: 2012-07-20
Senior Member
Posts: 9837
Joined: 2012-07-20
#5531580 Posted on: 03/26/2018 05:56 AM
Usual Suspects
Geekbench 3/4, Luxmark, Cinebench r15, 3D Mark FS Standard. And if you have some game with Benchmark built-in.
I sorted them out based on expected ability to show difference.
I can help with that. Any specific tests to run?
Usual Suspects
Geekbench 3/4, Luxmark, Cinebench r15, 3D Mark FS Standard. And if you have some game with Benchmark built-in.
I sorted them out based on expected ability to show difference.
thatguy91
Senior Member
Posts: 6648
Joined: 2010-08-27
Senior Member
Posts: 6648
Joined: 2010-08-27
#5531585 Posted on: 03/26/2018 06:53 AM
If they did have an 8 core, maybe they did the sensible thing and got rid of the integrated GPU. A very considerable number of people with Intel processors use them for gaming, so they get the higher spec CPU's. They then use a discrete graphics card and the GPU goes to waste. For businesses, they tend to buy the 'lesser' processors, those that gamers won't buy, so integrated GPU's for them makes sense.
I don't see the need for yet another chipset to support these CPU's, I guess it's so people with 6-core Coffee Lakes buy yet another motherboard to make the most of it. Also, there's the probability that the CPU will be an okay-ish price considering, but they'll make that up with the cost of the Z390 boards.
If this release is true, it probably means Ice Lake won't be until next year. This in all likelihood is what will happen due to Intel's ongoing issues with the 10 nm node. Therefore, Ice Lake will probably go up against Zen 2. Zen 2 is the first major upgrade to Zen, and is the successor to the yet to be released Zen+. The Zen 2 will supposedly be on a 7 nm node.
If they did have an 8 core, maybe they did the sensible thing and got rid of the integrated GPU. A very considerable number of people with Intel processors use them for gaming, so they get the higher spec CPU's. They then use a discrete graphics card and the GPU goes to waste. For businesses, they tend to buy the 'lesser' processors, those that gamers won't buy, so integrated GPU's for them makes sense.
I don't see the need for yet another chipset to support these CPU's, I guess it's so people with 6-core Coffee Lakes buy yet another motherboard to make the most of it
. Also, there's the probability that the CPU will be an okay-ish price considering, but they'll make that up with the cost of the Z390 boards.If this release is true, it probably means Ice Lake won't be until next year. This in all likelihood is what will happen due to Intel's ongoing issues with the 10 nm node. Therefore, Ice Lake will probably go up against Zen 2. Zen 2 is the first major upgrade to Zen, and is the successor to the yet to be released Zen+. The Zen 2 will supposedly be on a 7 nm node.
Koniakki
Senior Member
Posts: 2843
Joined: 2009-09-15
Senior Member
Posts: 2843
Joined: 2009-09-15
#5531846 Posted on: 03/26/2018 07:08 PM
Usual Suspects
Geekbench 3/4, Luxmark, Cinebench r15, 3D Mark FS Standard. And if you have some game with Benchmark built-in.
I sorted them out based on expected ability to show difference.
All done. Run/tested multiple time to eliminate the variations as much possible.
Also checked with a few other program/benchmarks but difference was either none or negligible.
So, some of the results below I would consider them a "best case" for pointing out differences.
But, in others, yeah, they are clearly there.
P.S: And excuse my horrendous table/format results. :p
DDR4 3200MHz CL15 vs 2400MHz CL15
https://browser.geekbench.com/v4/cpu/compare/7636528?baseline=7637463
DDR4 4266MHz CL17 vs 2400MHz CL15
https://browser.geekbench.com/v4/cpu/compare/7639075?baseline=7637463
2400CL15: 6232/21537 --- 6232/21314
3200CL15: 6555/22792 --- 6555/22770
Usual Suspects
Geekbench 3/4, Luxmark, Cinebench r15, 3D Mark FS Standard. And if you have some game with Benchmark built-in.
I sorted them out based on expected ability to show difference.
All done. Run/tested multiple time to eliminate the variations as much possible.
Also checked with a few other program/benchmarks but difference was either none or negligible.
So, some of the results below I would consider them a "best case" for pointing out differences.
But, in others, yeah, they are clearly there.
P.S: And excuse my horrendous table/format results. :p
+---------------------+--------------------------------------------------------------+
| | Intel i7 8700K@5.0/4.6 |
+---------------------+--------------------------------------------------------------+
| | 2400MHz | 3200MHz | 4266MHz |
| | 15-15-15-30-360 2T | 15-15-15-30-360 2T | 17-17-17-38-360 2T |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| Aida64 Memory | | | |
+---------------------+--------------------+--------------------+--------------------+
| Read | 37708 | 50049 | 63737 |
+---------------------+--------------------+--------------------+--------------------+
| Write | 37353 | 49949 | 65343 |
+---------------------+--------------------+--------------------+--------------------+
| Copy | 31986 | 43090 | 56438 |
+---------------------+--------------------+--------------------+--------------------+
| Latency | 51.1 | 41.6 | 37.3 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| Geekbench 4.2.2 | | | |
+---------------------+--------------------+--------------------+--------------------+
| ST | 6211 | 6551 | 6807 |
+---------------------+--------------------+--------------------+--------------------+
| MT | 27287 | 29595 | 31193 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| LuxMark 3.1 C++ | | | |
+---------------------+--------------------+--------------------+--------------------+
| LuxBall HDR | 5082 | 5290 | 5490 |
+---------------------+--------------------+--------------------+--------------------+
| Hotel | 1371 | 1385 | 1415 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| RealBench 2.44 | | | |
+---------------------+--------------------+--------------------+--------------------+
| Encoding | 271,877 | 281,477 | 291,087 |
+---------------------+--------------------+--------------------+--------------------+
| Heavy Multitasking | 222,266 | 249,176 | 267,120 |
+---------------------+--------------------+--------------------+--------------------+
| System Score | 186,288 | 195,848 | 203,071 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| Fritz Chess Bench | 28508 | 29298 | 29519 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| Rise Of Tomb Raider | | | |
+---------------------+--------------------+--------------------+--------------------+
| 720p VH/SMAA | 201 | 221 | 226 |
+---------------------+--------------------+--------------------+--------------------+
| 1080p VH/SMAA | 168.56 | 173.7 | 177.7 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| x265 v2.1.0 1080p | 53.83 | 55.84 | 56,82 |
+---------------------+--------------------+--------------------+--------------------+
| | Intel i7 8700K@5.0/4.6 |
+---------------------+--------------------------------------------------------------+
| | 2400MHz | 3200MHz | 4266MHz |
| | 15-15-15-30-360 2T | 15-15-15-30-360 2T | 17-17-17-38-360 2T |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| Aida64 Memory | | | |
+---------------------+--------------------+--------------------+--------------------+
| Read | 37708 | 50049 | 63737 |
+---------------------+--------------------+--------------------+--------------------+
| Write | 37353 | 49949 | 65343 |
+---------------------+--------------------+--------------------+--------------------+
| Copy | 31986 | 43090 | 56438 |
+---------------------+--------------------+--------------------+--------------------+
| Latency | 51.1 | 41.6 | 37.3 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| Geekbench 4.2.2 | | | |
+---------------------+--------------------+--------------------+--------------------+
| ST | 6211 | 6551 | 6807 |
+---------------------+--------------------+--------------------+--------------------+
| MT | 27287 | 29595 | 31193 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| LuxMark 3.1 C++ | | | |
+---------------------+--------------------+--------------------+--------------------+
| LuxBall HDR | 5082 | 5290 | 5490 |
+---------------------+--------------------+--------------------+--------------------+
| Hotel | 1371 | 1385 | 1415 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| RealBench 2.44 | | | |
+---------------------+--------------------+--------------------+--------------------+
| Encoding | 271,877 | 281,477 | 291,087 |
+---------------------+--------------------+--------------------+--------------------+
| Heavy Multitasking | 222,266 | 249,176 | 267,120 |
+---------------------+--------------------+--------------------+--------------------+
| System Score | 186,288 | 195,848 | 203,071 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| Fritz Chess Bench | 28508 | 29298 | 29519 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| Rise Of Tomb Raider | | | |
+---------------------+--------------------+--------------------+--------------------+
| 720p VH/SMAA | 201 | 221 | 226 |
+---------------------+--------------------+--------------------+--------------------+
| 1080p VH/SMAA | 168.56 | 173.7 | 177.7 |
+---------------------+--------------------+--------------------+--------------------+
| | | | |
+---------------------+--------------------+--------------------+--------------------+
| x265 v2.1.0 1080p | 53.83 | 55.84 | 56,82 |
+---------------------+--------------------+--------------------+--------------------+
DDR4 3200MHz CL15 vs 2400MHz CL15
https://browser.geekbench.com/v4/cpu/compare/7636528?baseline=7637463
DDR4 4266MHz CL17 vs 2400MHz CL15
https://browser.geekbench.com/v4/cpu/compare/7639075?baseline=7637463
Intel 8700K@4C/8T
2400CL15: 6232/21537 --- 6232/21314
3200CL15: 6555/22792 --- 6555/22770
Fox2232
Senior Member
Posts: 9837
Joined: 2012-07-20
Senior Member
Posts: 9837
Joined: 2012-07-20
#5531999 Posted on: 03/27/2018 05:56 AM
All done. Run/tested multiple time to eliminate the variations as much possible.
Also checked with a few other program/benchmarks but difference was either none or negligible.
So, some of the results below I would consider them a "best case" for pointing out differences.
But, in others, yeah, they are clearly there.
DDR4 3200MHz CL15 vs 2400MHz CL15
https://browser.geekbench.com/v4/cpu/compare/7636528?baseline=7637463
DDR4 4266MHz CL17 vs 2400MHz CL15
https://browser.geekbench.com/v4/cpu/compare/7639075?baseline=7637463
Great Work. Geekbench is most interesting. Some of workloads are practically unaffected, and some show even 25% scaling up from 2400 to 3200MHz.
Those which did not scale up are workloads which did not hit memory bandwidth limitation.
Those which did scale up require good bandwidth. Now I wonder for one last test:
May you please disable 2cores in BIOS? And run 2400/3200MHz memory + 4C/8T as controlling sample. (Geekbench only)
Looking at ST SGEMM which is in either case 133GFlops, and then MT scales with bandwidth to:
2400MHz - 504
3200MHz - 630
4266MHz - 710
I wonder if just 4C/8T drop it to theoretical 133*4 = 532. (Maybe geekbench 4 started to support SMT, maybe not.)
All done. Run/tested multiple time to eliminate the variations as much possible.
Also checked with a few other program/benchmarks but difference was either none or negligible.
So, some of the results below I would consider them a "best case" for pointing out differences.
But, in others, yeah, they are clearly there.
╔═════════════════════╦══════════════════════════════════════════════════════════════╗
║ ║ Intel i7 8700K@5.0/4.6 ║
╠═════════════════════╬════════════════════╦════════════════════╦════════════════════╣
║ ║ 2400MHz ║ 3200MHz ║ 4266MHz ║
║ ║ 15-15-15-30-360 2T ║ 15-15-15-30-360 2T ║ 17-17-17-38-360 2T ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Aida64 Memory ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Read ║ 37708 ║ 50049 ║ 63737 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Write ║ 37353 ║ 49949 ║ 65343 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Copy ║ 31986 ║ 43090 ║ 56438 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Latency ║ 51.1 ║ 41.6 ║ 37.3 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Geekbench 4.2.2 ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ST ║ 6211 ║ 6551 ║ 6807 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ MT ║ 27287 ║ 29595 ║ 31193 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ LuxMark 3.1 C++ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ LuxBall HDR ║ 5082 ║ 5290 ║ 5490 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Hotel ║ 1371 ║ 1385 ║ 1415 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ RealBench 2.44 ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Encoding ║ 271,877 ║ 281,477 ║ 291,087 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Heavy Multitasking ║ 222,266 ║ 249,176 ║ 267,120 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ System Score ║ 186,288 ║ 195,848 ║ 203,071 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Fritz Chess Bench ║ 28508 ║ 29298 ║ 29519 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Rise Of Tomb Raider ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ 720p VH/SMAA ║ 201 ║ 221 ║ 226 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ 1080p VH/SMAA ║ 168.56 ║ 173.7 ║ 177.7 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ x265 v2.1.0 1080p ║ 53.83 ║ 55.84 ║ 56.82 ║
╚═════════════════════╩════════════════════╩════════════════════╩════════════════════╝
║ ║ Intel i7 8700K@5.0/4.6 ║
╠═════════════════════╬════════════════════╦════════════════════╦════════════════════╣
║ ║ 2400MHz ║ 3200MHz ║ 4266MHz ║
║ ║ 15-15-15-30-360 2T ║ 15-15-15-30-360 2T ║ 17-17-17-38-360 2T ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Aida64 Memory ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Read ║ 37708 ║ 50049 ║ 63737 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Write ║ 37353 ║ 49949 ║ 65343 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Copy ║ 31986 ║ 43090 ║ 56438 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Latency ║ 51.1 ║ 41.6 ║ 37.3 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Geekbench 4.2.2 ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ST ║ 6211 ║ 6551 ║ 6807 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ MT ║ 27287 ║ 29595 ║ 31193 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ LuxMark 3.1 C++ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ LuxBall HDR ║ 5082 ║ 5290 ║ 5490 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Hotel ║ 1371 ║ 1385 ║ 1415 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ RealBench 2.44 ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Encoding ║ 271,877 ║ 281,477 ║ 291,087 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Heavy Multitasking ║ 222,266 ║ 249,176 ║ 267,120 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ System Score ║ 186,288 ║ 195,848 ║ 203,071 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Fritz Chess Bench ║ 28508 ║ 29298 ║ 29519 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ Rise Of Tomb Raider ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ 720p VH/SMAA ║ 201 ║ 221 ║ 226 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ 1080p VH/SMAA ║ 168.56 ║ 173.7 ║ 177.7 ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ ║ ║ ║ ║
╠═════════════════════╬════════════════════╬════════════════════╬════════════════════╣
║ x265 v2.1.0 1080p ║ 53.83 ║ 55.84 ║ 56.82 ║
╚═════════════════════╩════════════════════╩════════════════════╩════════════════════╝
DDR4 3200MHz CL15 vs 2400MHz CL15
https://browser.geekbench.com/v4/cpu/compare/7636528?baseline=7637463
DDR4 4266MHz CL17 vs 2400MHz CL15
https://browser.geekbench.com/v4/cpu/compare/7639075?baseline=7637463
Great Work. Geekbench is most interesting. Some of workloads are practically unaffected, and some show even 25% scaling up from 2400 to 3200MHz.
Those which did not scale up are workloads which did not hit memory bandwidth limitation.
Those which did scale up require good bandwidth. Now I wonder for one last test:
May you please disable 2cores in BIOS? And run 2400/3200MHz memory + 4C/8T as controlling sample. (Geekbench only)
Looking at ST SGEMM which is in either case 133GFlops, and then MT scales with bandwidth to:
2400MHz - 504
3200MHz - 630
4266MHz - 710
I wonder if just 4C/8T drop it to theoretical 133*4 = 532. (Maybe geekbench 4 started to support SMT, maybe not.)
jaggerwild
Senior Member
Posts: 709
Joined: 2005-11-18
Senior Member
Posts: 709
Joined: 2005-11-18
#5532012 Posted on: 03/27/2018 07:03 AM
Stop highjacking the thread, with your AMD blow fish lips! Post your numbers when they get a CPU, highjack that thread no mods in here?
Stop highjacking the thread, with your AMD blow fish lips! Post your numbers when they get a CPU, highjack that thread no mods in here?
Koniakki
Senior Member
Posts: 2843
Joined: 2009-09-15
Senior Member
Posts: 2843
Joined: 2009-09-15
#5532184 Posted on: 03/27/2018 02:31 PM
Great Work. Geekbench is most interesting. Some of workloads are practically unaffected, and some show even 25% scaling up from 2400 to 3200MHz.
Those which did not scale up are workloads which did not hit memory bandwidth limitation.
Those which did scale up require good bandwidth. Now I wonder for one last test:
May you please disable 2cores in BIOS? And run 2400/3200MHz memory + 4C/8T as controlling sample. (Geekbench only)
Looking at ST SGEMM which is in either case 133GFlops, and then MT scales with bandwidth to:
2400MHz - 504
3200MHz - 630
4266MHz - 710
I wonder if just 4C/8T drop it to theoretical 133*4 = 532. (Maybe geekbench 4 started to support SMT, maybe not.)
Results table in the previous post updated. A GB links also at the end for the 4C/8T.
At 2400MHz the results varied a bit, but I would say you were spot on.
Stop highjacking the thread, with your AMD blow fish lips! Post your numbers when they get a CPU, highjack that thread no mods in here?
Great Work. Geekbench is most interesting. Some of workloads are practically unaffected, and some show even 25% scaling up from 2400 to 3200MHz.
Those which did not scale up are workloads which did not hit memory bandwidth limitation.
Those which did scale up require good bandwidth. Now I wonder for one last test:
May you please disable 2cores in BIOS? And run 2400/3200MHz memory + 4C/8T as controlling sample. (Geekbench only)
Looking at ST SGEMM which is in either case 133GFlops, and then MT scales with bandwidth to:
2400MHz - 504
3200MHz - 630
4266MHz - 710
I wonder if just 4C/8T drop it to theoretical 133*4 = 532. (Maybe geekbench 4 started to support SMT, maybe not.)
Results table in the previous post updated. A GB links also at the end for the 4C/8T.
At 2400MHz the results varied a bit, but I would say you were spot on.
Stop highjacking the thread, with your AMD blow fish lips! Post your numbers when they get a CPU, highjack that thread no mods in here?
Noisiv
Senior Member
Posts: 6664
Joined: 2010-11-16
Senior Member
Posts: 6664
Joined: 2010-11-16
#5532635 Posted on: 03/28/2018 01:15 PM
If they did have an 8 core, maybe they did the sensible thing and got rid of the integrated GPU. A very considerable number of people with Intel processors use them for gaming, so they get the higher spec CPU's. They then use a discrete graphics card and the GPU goes to waste. For businesses, they tend to buy the 'lesser' processors, those that gamers won't buy, so integrated GPU's for them makes sense.
I like my Intel HD
Total power-in of my PC, while idling on desktop, browsing Guru3D.
75W on Intel HD (via HDMI input/output cable)
125W on R9 290 (via display port 1.2) (mem goes between 300-1250MHz, but this doesn't seem to be culprit)
------------------
That's cool 50W difference just one click away:
If they did have an 8 core, maybe they did the sensible thing and got rid of the integrated GPU. A very considerable number of people with Intel processors use them for gaming, so they get the higher spec CPU's. They then use a discrete graphics card and the GPU goes to waste. For businesses, they tend to buy the 'lesser' processors, those that gamers won't buy, so integrated GPU's for them makes sense.
I like my Intel HD
Total power-in of my PC, while idling on desktop, browsing Guru3D.
75W on Intel HD (via HDMI input/output cable)
125W on R9 290 (via display port 1.2) (mem goes between 300-1250MHz, but this doesn't seem to be culprit)
------------------
That's cool 50W difference just one click away:
Fox2232
Senior Member
Posts: 9837
Joined: 2012-07-20
Senior Member
Posts: 9837
Joined: 2012-07-20
#5532806 Posted on: 03/28/2018 08:44 PM
Results table in the previous post updated. A GB links also at the end for the 4C/8T.
At 2400MHz the results varied a bit, but I would say you were spot on.
Thanks. Tiny Variation does not matter at all. Differences between each configurations are incomparably bigger.
https://browser.geekbench.com/v4/cpu/compare/7655359?baseline=7637463
Left: 4C/8T + 3200MHz memory ; Right: 6C/12T + 2400MHz memory.
Shows how starved even 6C/12T can be with 2400MHz. I did series of calculations and following are results:
Approximated based on performance scaling from 4C/8T+3200MHz to 6C/12T+2400MHz; 6C/12T+3200MHz; 6C/12T+4266MHz. While considering average 35% scaling as target (instead of 50%).
Results table in the previous post updated. A GB links also at the end for the 4C/8T.
At 2400MHz the results varied a bit, but I would say you were spot on.
Thanks. Tiny Variation does not matter at all. Differences between each configurations are incomparably bigger.
https://browser.geekbench.com/v4/cpu/compare/7655359?baseline=7637463
Left: 4C/8T + 3200MHz memory ; Right: 6C/12T + 2400MHz memory.
Shows how starved even 6C/12T can be with 2400MHz. I did series of calculations and following are results:
Approximated based on performance scaling from 4C/8T+3200MHz to 6C/12T+2400MHz; 6C/12T+3200MHz; 6C/12T+4266MHz. While considering average 35% scaling as target (instead of 50%).
Memory clock to extract close to maximum in given Workload from all threads.
6C/12T 8C/16T
AES 4898 6531
LZMA 2863 3818
JPEG 2643 3524
Canny 4422 5896
Lua 2221 2961
Dijkstra 4385 5847
SQLite 2579 3439
HTML5 Parse 2862 3816
HTML5 DOM 4562 6083
Histogram Equalization 3826 5101
PDF Rendering 4891 6521
LLVM 2390 3186
Camera 2852 3803
SGEMM 4161 5548
SFFT 2173 2898
N-Body Physics 2165 2887
Ray Tracing 2411 3215
Rigid Body Physics 2206 2941
HDR 4209 5613
Gaussian Blur 2863 3818
Speech Recognition 4552 6070
Face Detection 2184 2912
6C/12T 8C/16T
AES 4898 6531
LZMA 2863 3818
JPEG 2643 3524
Canny 4422 5896
Lua 2221 2961
Dijkstra 4385 5847
SQLite 2579 3439
HTML5 Parse 2862 3816
HTML5 DOM 4562 6083
Histogram Equalization 3826 5101
PDF Rendering 4891 6521
LLVM 2390 3186
Camera 2852 3803
SGEMM 4161 5548
SFFT 2173 2898
N-Body Physics 2165 2887
Ray Tracing 2411 3215
Rigid Body Physics 2206 2941
HDR 4209 5613
Gaussian Blur 2863 3818
Speech Recognition 4552 6070
Face Detection 2184 2912
Click here to post a comment for this news story on the message forum.
Member
Posts: 32
Joined: 2017-02-27
Ryzen and Epyc are different platforms, so stop comparing apples and oranges to get your points across. /benchmarks I saw tells the different story, depends on a number of threads, sometimes AMD is better, sometimes Intel, one AMD CCX /4cores/ can access only up to 30GB/s /
8700k and 1800x have pretty much same memory bandwith while 8700k having 3x lower latencies, also i7 8700k's cache manages higher bandwith, so not really. OC 8700k is almost as quick or even faster as an 8 core Ryzen in cpu heavy tasks such as rendering etc. Intel also generally handles and supports higher RAM clocks and lower latencies better.
None of current Intel cpus were developed as quad cores, all cores all native and the ringbus is scalable, the culprit is that at a certain point you start encountering higher latencies due to a higher number of cycles, so Intel is upgrading to the mesh.
If anything, AMD current cpus were designed mainly as quadcores which are interconnected together in order to be scaled, look at their latencies and their gaming performance being heavily affected by RAM clocks/timings. So stop with this projecting towards Intel.
So I really don't get it were are you getting your assumptions from because by your logic, even 8700k should be somehow underpeforming and suffering from "the quad core design" and "missing the quad channel". But the reality is different and 8700k is riping everything apart in games and contests with 8cores when OC-ed at rendering.