Associate
- Joined
- 27 Apr 2007
- Posts
- 966
In the graph above it consumes more than Intel's 65W part.
But that's a different discussion anyway as I was replying to a quote about the top of the range parts which are nowhere near 65W.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Core 9000 series
- Thread starter orbitalwalsh
- Start date
But that's a different discussion anyway as I was replying to a quote about the top of the range parts which are nowhere near 65W.
Soldato
The problem is that power figures tend to vary a lot depending on the review you read which is understandable as it can vary purely based on the software you are running.
So unless someone links to actual reviews I don't take it seriously.
I just looked at 2 reviews and got the info below.
These were the 1st two I looked at with power figures and I don't pretend to know if they are indicative as I prefer not to believe my own hype.
TechReport
Techpowerup
LOL. No overclocked 8700K there.....
Have a look here....
https://www.reddit.com/r/Amd/comments/8df2ob/look_at_that_power_consumption2700x_104w_vs/
And overclocking 8700K to 5Ghz requires a massive jump in power.
Here is a good example also
http://www.overclock.net/forum/5-in...-max-package-power-let-s-compile-results.html
- 8700k - 4.7GHz / 1.20v (delidded)
- ASRock Taichi
- Prime95 v29.4 AVX Small FFTs (5 mins)
- HWINFO CPU Package Power: 165 watts
- Max Core Temp: 67c
- 8700k - 5.0Ghz / 1.328v (delidded)
- ASRock Taichi
- Prime95 v29.4 AVX Small FFTs (5 mins)
- HWINFO CPU Package Power: 225 watts
- Max Core Temp: 77c
That is less than 8% overclock but a not linear +37% higher power consumption.
Again I am not one of those who says "duh need to save on power", but the argument was that overclocked 2700X with it's extra 2 cores is more power efficient than the 5Ghz 8700K and is faster on multi thread stuff.
And you can imagine how much the 8 CFL CPU going to burn to achieve 5Ghz. I doubt there is Z370 board being able to deliver 310W+ right now. I know for certain my Z370 wont.
And trying to move 310W+ without deliding and liquid metal, that will be a big feat.
Last edited:
Associate
- Joined
- 14 Oct 2004
- Posts
- 979
Power requirement isn't linear with core count tho is it?
If the 8700K consumes 225watts with 6 cores, it's not going to drop to 150watts with 4 cores.
The flipside being if you add 33% more cores (6 to 8) it won't need 33% more power.
Anyone know what the actual numbers are?
If the 8700K consumes 225watts with 6 cores, it's not going to drop to 150watts with 4 cores.
The flipside being if you add 33% more cores (6 to 8) it won't need 33% more power.
Anyone know what the actual numbers are?
Caporegime
Notice something about that. Ryzen 2700X TDP is 105 Watts, power consumption is 104.7 Watts... in fact all the AMD CPU are with in thier TDP ratings, even the FX series.
Now look at all the Intel CPU's, for example the 8700K has a TDP rating of 95 Watt, 10 Watts less than the 2700X, HA! how convenient Intel, actual power consumption, 159.5 Watts.
The 7900X, TDP 140 Watts, actual power consumption a ridiculous 229.6 watts, it has 2 more cores than the 2700X but uses way more than twice the amount of power.
Intel are extremely generous when citing their own TDP ratings, no wonder they all need enthusiast grade cooling and deliding.
Associate
- Joined
- 27 Apr 2007
- Posts
- 966
LOL. Over-clocking was never mentioned.
Caporegime
I'll say one thing, looking around at power consumption reviews, the Intel results vary MASSIVELY depending on where you look, by as much has 100 watts, while all the AMD results are consistent from one review to another, WTF is with that?
Are all reviewers AMD shills?
Are all reviewers AMD shills?
Last edited:
Caporegime
You know what this needs to be said, i'll give you an example of this, Toms Hardware have the system power consumption of the 7600K at just over 100 Watts, that's the CPU, the Chip-Set, the Memory Modules.......its also about the same as the 2700X. Tech Power Up have the 7600K full system at 70 watts, 70 watt for the full system doesn't leave much for the CPU, a pair of Memory Modules alone are 30 watts when stressed, which they are running Cinebench.
At the same time Tech Power up are saying the 2700X uses twice as much power than what Toms Hardware say it does.
I noticed this before, PCPer when they first reviewed SkyLake-X CPU's all of them according to them just so happened to fit neatly inside their TDP ratings, for example the 7900X was 139 Watts, 1 watt under Intel's TDP rating, when everyone else was publishing over 200 Watts.
When i ran Blender on my 4690K the power consumption was about 120 Watts at stock, close to 200 overclocked, the 1600 uses the same power at stock, 150 overclocked, Tech Power Up i don't believe you.
When someone like Intel sends you review advice and is clear about what they power consumption of their CPU's is you don't want contradict them if you want to remain their partners.
Soldato
I just noticed guru3d has the Vega 64 as running with 334W when gaming... I only see that when benching at 1748 core. Usually is around 250-260 (no downvoltimg just factory overclock 1630 on turbo mode)
The opposite applies to 1080Ti. The Extreme was over 320W constantly when gaming
Soldato
IPC is now within a few percent, less than 5%.
The 9*** series you'd think is a new setup. Naming is another notch "higher" on the roll you'd think....when in fact it's just another rebrand. Yet INTEL are the greatest and whatever they do, people will blindly follow them. Like apple fanboys and their iPhones.
Not YOU specifically Silent, just figuratively speaking.
Soldato
IMC with 2000 series is not issue. If your memory module is not dual band hynix, it will ran at its rated speed.
Some had no issues to put 4000Mhz ram on the 2700X and run it. My 1800X could easily work with 3600Mhz ram on the CH6 with 0701 bios. (native ram speed)
Also good you brought this up about the IPC.
http://www.guru3d.com/articles_pages/amd_ryzen_7_2700x_review,9.html
So at same speed the difference between Zen+ (2000 series) and Coffee Lake is 0.026%.
It just shows that the IPC of Zen+ is equal to Skylake. Imho not a bad feat given is a brand new architecture. (Zen 1 in total).
The only thing that AMD doesn't have with Zen+ is turbo speeds but these are coming with Zen 2 next year.
Also the performance of the Zen CPUs is directly related to RAM.
Is a bit extensive review but they do damn good job. (use google translate)
https://www.computerbase.de/2018-04/amd-ryzen-2000-test/7/#diagramm-total-war-warhammer-uebertakten
As you shall see the 10% higher boost speed of the 8700K goes out of the window, if both are using the same ram (3466 tight timings).
And yes 8700K benefits greatly from fast tight ram as clearly shown there also.
But not as great as 20% FPS boost from 2933 to 3466 the Ryzen+ enjoys.
Caporegime
And Guru3D like a lot of other people assume that single threaded IPC translates directly to multithreaded IPC, of course it doesn't, AMD Multithreaded scaling is better than Intel's.
So with that at 4Ghz the 12 thread 2600X scored 1384, the 8700K at the same speed scored 1325, that makes the IPC in this instant 4.5% higher on Ryzen 2.
Even single threaded the 8700K is only 3.5% ahead.
In both instances they are close enough to eachother. So I think this argument that Intel have higher IPC is now false, they just don't. not even when it comes to gaming, clock for clock there is no difference overall, its give and take depending on the game while no one is ever largely different.
Soldato
And Guru3D like a lot of other people assume that single threaded IPC translates directly to multithreaded IPC, of course it doesn't, AMD Multithreaded scaling is better than Intel's.
So with that at 4Ghz the 12 thread 2600X scored 1384, the 8700K at the same speed scored 1325, that makes the IPC in this instant 4.5% higher on Ryzen 2.
Even single threaded the 8700K is only 3.5% ahead.
In both instances they are close enough to eachother. So I think this argument that Intel have higher IPC is now false, they just don't. not even when it comes to gaming, clock for clock there is no difference overall, its give and take depending on the game while no one is ever largely different.
I do not disagree that IPC means nothing especially when RAM & motherboard play important parts and multi core performance scales differently.
See the link from the German review site how differently the 2700X behaves when latency drops from 74ns to 60ns.
Caporegime
Yup, i did see that, i did something similar, tho obviously nothing like as well as they did....
https://forums.overclockers.co.uk/threads/ryzen-1-memory-scaling-testing.18813342/
Soldato
It's not extensive, really. If you're comparing latency and frequency but not sub timings, then you're covering already tread ground. All depends on how much one wants to tune, however, Zen+ still isn't capable of running timings at or close to the minimum spacing required by the chipset, something that's been viable on Intel's part since Broadwell-E. Something that's especially impressive when nearing 3866-4000Mhz on SKL and beyond.
Unless one wants to run Cinebench all day, the synthetic cherry-picking only goes so far, especially when it comes to gaming.
Of course the intel thread is 95% Ryzen 
Particularly enjoyed the bit about 4,000mhz memory on Ryzen 2x where ‘some’ had issues. I call BS on it being anywhere close to compatible with that level of high memory speeds, it just isn’t true - I’ve owned an 1800x and now 8700k and the pain people go through just tweaking Ryzen to work with memory is incredible. Compare that to the 8700k where I selected XMP and it worked flawlessly at 4,133 instead of scraping 3,600 with pretty terrible timings in a bios released 6 months after the product (thankfully because one genius (Elmor) was tirelessly working on it).
Anyway, 9xxx looks like a miss given my 8700k will run 5.3 doesn’t seem likely it’s going to be worth while... But you never know!
Particularly enjoyed the bit about 4,000mhz memory on Ryzen 2x where ‘some’ had issues. I call BS on it being anywhere close to compatible with that level of high memory speeds, it just isn’t true - I’ve owned an 1800x and now 8700k and the pain people go through just tweaking Ryzen to work with memory is incredible. Compare that to the 8700k where I selected XMP and it worked flawlessly at 4,133 instead of scraping 3,600 with pretty terrible timings in a bios released 6 months after the product (thankfully because one genius (Elmor) was tirelessly working on it).
Anyway, 9xxx looks like a miss given my 8700k will run 5.3 doesn’t seem likely it’s going to be worth while... But you never know!
Soldato
Intel's memory controller is leagues better than Zen. But then it was designed in-house, too. Zen+ is what Ryzen should have been last year. It wasn't fit for release, and the ucode is now, a year after release, fit for public consumption (if wanting to use overclocked memory kits).
In AMD's defense, there was always going to be a period where there would be a large number of users assuming that kits validated for other platforms would work, purely on the basis that AMD was touting 3200Mhz and beyond. DRAM overclocking is a complex subject that even a large portion of seasoned enthusiasts seldom understand.
In AMD's defense, there was always going to be a period where there would be a large number of users assuming that kits validated for other platforms would work, purely on the basis that AMD was touting 3200Mhz and beyond. DRAM overclocking is a complex subject that even a large portion of seasoned enthusiasts seldom understand.
Soldato
These issues existed for decade. Even 3000 series is affected.As i only browse the web and watch a few tv shows i'm thinking of going back to a cheap 6th or 7th gen low power build until they resolve all these spectre meltdown issues , Am i mad?
As for low power define it please. A 6700K, 7700K burn more power than the 8700K.
Exactly , so as they all have it i'm thinking of going in the opposite direction until it's hardware fixed , why waste more money on something that's already broken out of the box?
Memory plays fine on Ryzen 2 and X470. I've read of scant issues on the new boards. Ripping AMD for this is genuinely hilarious in light of Intel's shenanigans. The crappy IHS alone is just a joke... having to void your warranty just to get desirable temps is pathetic. Or of course pay a fortune for the likes of OCUK to do it for you, which makes even less sense.
Intel still have the edge in games but an 'edge' is all it is, you won't even notice in the vast majority of cases, especially above 1080p. I'm just fed up with their business practice and treating consumers like mugs, simply because they've had the market to themselves for so long. My switch to Ryzen 2 was an easy one personally, as it's been for many.
Intel still have the edge in games but an 'edge' is all it is, you won't even notice in the vast majority of cases, especially above 1080p. I'm just fed up with their business practice and treating consumers like mugs, simply because they've had the market to themselves for so long. My switch to Ryzen 2 was an easy one personally, as it's been for many.