AMD Ryzen 16 Core “Whitehaven” Enthusiast CPUs Leaked – 3.6GHz Clock Speed, Boatloads of Cache & Qua

[juanrga]

source?

Take the performance gap per core measured by AMD and by leaks reported in the former page. Then use the equation that gives mutithreaded performance, solve it and you get the crossing point is about 26 threads more or less.

 

[KazeoHin]

Take the performance gap per core measured by AMD and by leaks reported in the former page. Then use the equation that gives mutithreaded performance, solve it and you get the crossing point is about 26 threads more or less.

So, speculation stated as fact.

Got it.

 

[juanrga]

So, speculation stated as fact.

Got it.

It is robust estimation based in applying available data to the basic laws of compute.

 

[Osjur]

Repeating the same operation the CB score gives a ~20% performance gap per core

Yep, around 20% difference when comparing against stock 7900X @ 4Ghz all core turbo vs stock TR 1950X @ 3.5Ghz all core turbo. But when we compare against 16c 7960X, performance gap drops to ~9%. And as we know, there is no such thing as 16c / 32t Core i9 with 4.0Ghz stock all core turbo.

Here's some really rough math where core scaling penalty for TR 1950X is already accounted for (doh...) and scaling is perfect for i9 7960X. I guess I should use 1.55-58 as multiplier instead to get more accurate guess but nah, lets give Intel linear scaling. CB scores come from last page.

Stock scores:                                          all core turbo
i9 7900X Score: 2200p                              | 4.0ghz (10c / 20t)
i9 7960X Score: 3168p = 2200p * 1.6 * (3.6/4.0)    | 3.6ghz (16c / 32t)
TR 1950X Score: 2900p = 3400p * (3.5/4.1)          | 3.5ghz (16c / 32t)

3168p / 2900p = 7960X 9% faster vs TR1950X

For instance for 24 threads loaded on both chips. Intel would be faster. And all this is about comparing both chips on stock settings.

That is quite bold claim. 7900X doesn't have have the capacity to fully utilize 24 threads. it chokes after 20 threads if they are executed at full load at the same time. You really should know that adding more threads only hampers performance if the CPU doesn't have enough threads to execute them concurrently and they just start to stall more because of thread swapping.

Using more rough math and CB as example:

Threadripper
16t cores at 3.5Ghz: 2900 * 0.8 =2320p
1t SMT: (2900 - 2320) / 16 =~ 36p
16c + 8 SMT = 2320 + 36 * 8 = 2608

24t load: 2200p < 2600p

In reality that 2200 score would be a bit lower because 7900X can't magically shit out moar cores to increase performance.

Here's my bold claim: 16c / 16t TR is going to be faster than 10c / 20t 7900X on many workloads.

 

[arestavo]

Yep, around 20% difference when comparing against stock 7900X @ 4Ghz all core turbo vs stock TR 1950X @ 3.5Ghz all core turbo. But when we compare against 16c 7960X, performance gap drops to ~9%. And as we know, there is no such thing as 16c / 32t Core i9 with 4.0Ghz stock all core turbo.

Here's some really rough math where core scaling penalty for TR 1950X is already accounted for (doh...) and scaling is perfect for i9 7960X. I guess I should use 1.55-58 as multiplier instead to get more accurate guess but nah, lets give Intel linear scaling. CB scores come from last page.

Stock scores:                                          all core turbo
i9 7900X Score: 2200p                              | 4.0ghz (10c / 20t)
i9 7960X Score: 3168p = 2200p * 1.6 * (3.6/4.0)    | 3.6ghz (16c / 32t)
TR 1950X Score: 2900p = 3400p * (3.5/4.1)          | 3.5ghz (16c / 32t)

3168p / 2900p = 7960X 9% faster vs TR1950X


That is quite bold claim. 7900X doesn't have have the capacity to fully utilize 24 threads. it chokes after 20 threads if they are executed at full load at the same time. You really should know that adding more threads only hampers performance if the CPU doesn't have enough threads to execute them concurrently and they just start to stall more because of thread swapping.

Using more rough math and CB as example:

Threadripper
16t cores at 3.5Ghz: 2900 * 0.8 =2320p
1t SMT: (2900 - 2320) / 16 =~ 36p
16c + 8 SMT = 2320 + 36 * 8 = 2608

24t load: 2200p < 2600p

In reality that 2200 score would be a bit lower because 7900X can't magically shit out moar cores to increase performance.

Here's my bold claim: 16c / 16t TR is going to be faster than 10c / 20t 7900X on many workloads.

Am I reading it right that you're saying that the 16C 32T i9 will only theoretically get a 2200 CB15 multithread score? I personally find that funny, as my overclocked 6950X gets more than that.

As with all things - time will tell.

 

[tangoseal]

in a few hours all will be known then we can stop reading this conjecturing bicker-ness back and forth lol

 

[OrangeKhrush]

Yep, around 20% difference when comparing against stock 7900X @ 4Ghz all core turbo vs stock TR 1950X @ 3.5Ghz all core turbo. But when we compare against 16c 7960X, performance gap drops to ~9%. And as we know, there is no such thing as 16c / 32t Core i9 with 4.0Ghz stock all core turbo.

Here's some really rough math where core scaling penalty for TR 1950X is already accounted for (doh...) and scaling is perfect for i9 7960X. I guess I should use 1.55-58 as multiplier instead to get more accurate guess but nah, lets give Intel linear scaling. CB scores come from last page.

Stock scores:                                          all core turbo
i9 7900X Score: 2200p                              | 4.0ghz (10c / 20t)
i9 7960X Score: 3168p = 2200p * 1.6 * (3.6/4.0)    | 3.6ghz (16c / 32t)
TR 1950X Score: 2900p = 3400p * (3.5/4.1)          | 3.5ghz (16c / 32t)

3168p / 2900p = 7960X 9% faster vs TR1950X


That is quite bold claim. 7900X doesn't have have the capacity to fully utilize 24 threads. it chokes after 20 threads if they are executed at full load at the same time. You really should know that adding more threads only hampers performance if the CPU doesn't have enough threads to execute them concurrently and they just start to stall more because of thread swapping.

Using more rough math and CB as example:

Threadripper
16t cores at 3.5Ghz: 2900 * 0.8 =2320p
1t SMT: (2900 - 2320) / 16 =~ 36p
16c + 8 SMT = 2320 + 36 * 8 = 2608

24t load: 2200p < 2600p

In reality that 2200 score would be a bit lower because 7900X can't magically shit out moar cores to increase performance.

Here's my bold claim: 16c / 16t TR is going to be faster than 10c / 20t 7900X on many workloads.

Clock for clock analytics seem to avoid this one when he so chooses to turn the blind eye to it. I am actually shocked he even overlooked such a major factor, oh wait sorry, nothing shocks me more from this one anymore. I have become anesthetized by the cherry picker that calls others cherry pickers.

10% gap has been pretty commonly accepted in general compute standards for some time, even by AMD users. Given how noncompetitive AMD was with Bulldozer uARCH this is quite impressive, but of course some will tell you otherwise. His early projections were like 30-40% gaps so 10% to Chipzilla's performance is a good showing.

 

[OrangeKhrush]

Am I reading it right that you're saying that the 16C 32T i9 will only theoretically get a 2200 CB15 multithread score? I personally find that funny, as my overclocked 6950X gets more than that.

As with all things - time will tell.

it has a very low base and turbo so yes it is possible, by comparing an overclocked 6950X likely around 4.5Ghz it kind of makes this funny

 

[juanrga]

10% gap has been pretty commonly accepted in general compute standards for some time, even by AMD users.

It doesn't matter what certain users "accept", but what is objectively measured.

Given how noncompetitive AMD was with Bulldozer uARCH this is quite impressive, but of course some will tell you otherwise.

Everyone agrees that Zen is a giant leap compared to Bulldozer. What some people reject is all the hype, the exaggerated claims, and the double standards.

His early projections were like 30-40% gaps so 10% to Chipzilla's performance is a good showing.

The accuracy of 'his' predictions was of 90% or so, but keep pretending otherwise...

 

[juanrga]

Here's my bold claim: 16c / 16t TR is going to be faster than 10c / 20t 7900X on many workloads.

The 16C TR will be about 30% faster (using your own math) than 10C SKL only on throughput-optimized workloads with large SMT yields and that scale up to 32 threads, such as CB [*]. The 16C TR will lose on everything else: throughput-optimized workloads that cannot use the extra 60% cores and have low SMT yields; latency-optimized workloads, and AVX 256--512 workloads.


[*] It doesn't look fortuitous to me that CB scores are being leaked, whereas other metrics aren't leaked. Just as it was not fortuitous that CB and CPU-Z scores were leaked for RyZen, but other metrics weren't, because reviews of RyZen demonstrated that CB and CPU-Z were optimal points.

 

[JustReason]

The 16C TR will be about 30% faster (using your own math) than 10C SKL only on throughput-optimized workloads with large SMT yields and that scale up to 32 threads, such as CB [*]. The 16C TR will lose on everything else: throughput-optimized workloads that cannot use the extra 60% cores and have low SMT yields; latency-optimized workloads, and AVX 256--512 workloads.


[*] It doesn't look fortuitous to me that CB scores are being leaked, whereas other metrics aren't leaked. Just as it was not fortuitous that CB and CPU-Z scores were leaked for RyZen, but other metrics weren't, because reviews of RyZen demonstrated that CB and CPU-Z were optimal points.

You should look at the review of EPYC. It doesn't lose at everything, actually it does well enough to top the charts in nearly all benches even using that 9mth build that was optimized for SKL you have been crying about for the other test.

 

[juanrga]

You should look at the review of EPYC. It doesn't lose at everything, actually it does well enough to top the charts in nearly all benches even using that 9mth build that was optimized for SKL you have been crying about for the other test.

At what point "everything else" was transcribed as everything?

The STH EPYC review pretty much summarizes my point about TR. The 64-core EPYC server matches a 36-core SKL Xeon on a compile workload and then beats the Xeon on a rendering workload thanks to having 78% moar cores.

 

[OrangeKhrush]

You should look at the review of EPYC. It doesn't lose at everything, actually it does well enough to top the charts in nearly all benches even using that 9mth build that was optimized for SKL you have been crying about for the other test.

No you know why I said cherry picking, choosing ones perceptions and convincing oneself it is reality is an entertaining way to go about this.

 

[OrangeKhrush]

At what point "everything else" was transcribed as everything?

The STH EPYC review pretty much summarizes my point about TR. The 64-core EPYC server matches a 36-core SKL Xeon on a compile workload and then beats the Xeon on a rendering workload thanks to having 78% moar cores.

So in the workloads it is catered towards it does well...gotcha, never thought that was a bad thing. I mean they don't need special compilers and AVX to handicap use to certain features and monopolies a segment. In domains where Intel vs AMD is treated evenly then AMD pulls out ahead far to many times for the boys in blue to like.

 

[arestavo]

it has a very low base and turbo so yes it is possible, by comparing an overclocked 6950X likely around 4.5Ghz it kind of makes this funny

All core turbo is 3.6GHz for the 16 core 7960X at stock, or 20% slower than what I've got my 10 core 6950X running at.

I know, math sucks (it really does) - but you're wrong on this and I've called Osjur on it.

And not to harp on you about the 6950X, but the average OC for it is 4.3GHz not 4.5. Mine does 4.4 (19 percentile).

All that said I want Threadripper to be kick ass and NOT be another Bulldozer. Intel needs some competition after stagnating all these years.

 

[juanrga]

So in the workloads it is catered towards it does well...gotcha, never thought that was a bad thing.

It does well on throughput-like tasks such as rendering and encoding, but we don't need reviews for that. Some of us predicted, before launch, it was going to shine on such tasks. Too bad those thoughput-like tasks often run better on alternative hardware including GPUs.

I mean they don't need special compilers and AVX to handicap use to certain features and monopolies a segment. In domains where Intel vs AMD is treated evenly then AMD pulls out ahead far to many times for the boys in blue to like.

Hum, customers seem very happy about using AVX. Many of them shared with us their experiences on how AVX improves performance and efficiency on their real-life workloads (in some cases by huge amounts as 2X). And of course no one is prohibiting AMD to make a CPU that shines on AVX workloads, which makes all your talk about monopolies even less relevant.

 

[FrgMstr]

AMD Threadripper Review is LIVE

 

[OrangeKhrush]

It does well on throughput-like tasks such as rendering and encoding, but we don't need reviews for that. Some of us predicted, before launch, it was going to shine on such tasks. Too bad those thoughput-like tasks often run better on alternative hardware including GPUs.



Hum, customers seem very happy about using AVX. Many of them shared with us their experiences on how AVX improves performance and efficiency on their real-life workloads (in some cases by huge amounts as 2X). And of course no one is prohibiting AMD to make a CPU that shines on AVX workloads, which makes all your talk about monopolies even less relevant.

Intel's very own AVX support page says that non Intel only get baseline AVX support, whatever baseline is it is significantly lower than full support. It is a propriatory standard and people are happy because that is all there was in the market at the time. AVX is catered to a market intel created by ensuring AMD could not sell enterprise grade products in the Opteron/Itanium era. AVX is purely a market standard by a sole player for a long time.

In general computing AVX is a bag of _____, so it is immaterial at this point.

 

[Osjur]

Am I reading it right that you're saying that the 16C 32T i9 will only theoretically get a 2200 CB15 multithread score? I personally find that funny, as my overclocked 6950X gets more than that.

As with all things - time will tell.

No, you are reading it wrong.

Stock clock scores with my math:
i9 7900X Score: 2200p
i9 7960X Score: 3168p (math to get this score: 2200p * 1.6 * (3.6/4.0)
TR 1950X Score: 2900p (math to get this score: 3400p * (3.5/4.1)

Second math part was for 7900X running against TR 1950X while both are utilizing 24 threads.

Oh well, my math was bit wrong, TR is actually getting little over 3000p in CB and its because I thought the all core turbo would be 3.5Ghz and I calculated that it would only get 2900p with that.

The real all core turbo for TR is 3.7Ghz.

EDIT: pic of the day (from one techsite):

Look how big the difference is

Kyle needs to make his graphs look like this.

 

[Gideon]

No, you are reading it wrong.

Stock clock scores with my math:
i9 7900X Score: 2200p
i9 7960X Score: 3168p (math to get this score: 2200p * 1.6 * (3.6/4.0)
TR 1950X Score: 2900p (math to get this score: 3400p * (3.5/4.1)

Second math part was for 7900X running against TR 1950X while both are utilizing 24 threads.

Oh well, my math was bit wrong, TR is actually getting little over 3000p in CB and its because I thought the all core turbo would be 3.5Ghz and I calculated that it would only get 2900p with that.

The real all core turbo for TR is 3.7Ghz.

EDIT: pic of the day (from one techsite):

Look how big the difference is

Kyle needs to make his graphs look like this.

Wow now that is once heck of a scale on that chart, could they not get it down to .01 frames?

 

[kac77]

No, you are reading it wrong.

Stock clock scores with my math:
i9 7900X Score: 2200p
i9 7960X Score: 3168p (math to get this score: 2200p * 1.6 * (3.6/4.0)
TR 1950X Score: 2900p (math to get this score: 3400p * (3.5/4.1)

Second math part was for 7900X running against TR 1950X while both are utilizing 24 threads.

Oh well, my math was bit wrong, TR is actually getting little over 3000p in CB and its because I thought the all core turbo would be 3.5Ghz and I calculated that it would only get 2900p with that.

The real all core turbo for TR is 3.7Ghz.

EDIT: pic of the day (from one techsite):

Look how big the difference is

Kyle needs to make his graphs look like this.

Did Juanrga make that?

 

[juanrga]

In general computing AVX is a bag of _____, so it is immaterial at this point.

AVX is an implementation of the SIMD-computation paradigm. SIMD-computation was never designed for general computing.

 

[juanrga]

No one here is discussing single-thread performance. Multitthread performance is given by the product of the performance of a single core and the number of cores. Intel having a faster core is the reason why AMD needs 60% moar cores to get 30% higher performance on workloads that scale to all the 16 cores.

On workloads that don't scale up to 32 threads, the TR chip will be slower because each core is slower. For instance for 24 threads loaded on both chips. Intel would be faster. And all this is about comparing both chips on stock settings. the i9 has higher overclocking headroom. Thus with both chips are overclocked the gap will increase from ~30% to ~45% and TR will be faster (~15%) only when all cores are fully loaded.

It seems my prediction Intel 10C chip would be faster than TR even under 24 threads workloads was rather good.

Also we see how both TR models are slower than the 1800X on games, as expected. I wait anxious to 1900X vs 1800X comparison.

 

[ecmaster76]

It seems my prediction Intel 10C chip would be faster than TR even under 24 threads workloads was rather good.

Also we see how both TR models are slower than the 1800X on games, as expected. I wait anxious to 1900X vs 1800X comparison.

TR is very impressive for a MCM/NUMA part. The fact that it is in the same league as Intel's monolithic mesh networked Xeon parts is quite remarkable.

I'd be curious to see how it compared to a 2 socket LCC Xeon system if they still make those

 

[Osjur]

It seems my prediction Intel 10C chip would be faster than TR even under 24 threads workloads was rather good.

No, It seems you didn't read the hardware.fr review. It actually proves my prediction and makes your look silly (and completely wrong). Yes, 7900X is tiny bit faster than 1920X when comparing across multiple programs, where some of the programs can't utilize all those threads properly.

Everytime the programs can fully utilize 24 threads, 1920X is either neck to neck with 7900X or ahead. 1950X is comfortably ahead in most cases where the program actually utilizes all available threads. And you are talking about workloads with up 24 threads, not SuperPI with single thread.

But your prediction was against 1950X with <24 threads. The thing is: 1920X which is pretty much neck to neck with 7900X, only has 12 cores and 24 threads. And I have a feeling that you really are treating all the threads on these cpu's as equal. Well, they're NOT.

If we compare 7900X to 1950X and 1920X with forced 24 thread utilization, we have 16 cores + 8 SMT threads on 1950X and 12 core + 12 SMT threads on 1920X. Now I'm gonna give you yet another newsflash. 16c / 16t imaginary TR is gonna be faster than 12c / 24t 1920X.

Its actually really easy to calculate and I'll do it for you: 1920X 12c * 1.2 (SMT avg. gain) = 14.4 cores. From that data we can extrapolate that TR with 16c / 16t is gonna be around 10% faster than 1920X, which also means that it beats 7900X even when comparing across multiple programs. Lets add those 8 extra SMT threads in to the mix and we have a TR with 16c 16t + 8 SMT which makes it ~20% faster than 1920X, and again faster than 7900X.

But I'm guessing this might be too much data for you to handle, so you can keep on believing that 7900X is gonna be faster than 1950X when running workloads up to 24 threads (even though 16c / 16t TR is gonna beat it when threads are utilized properly).

 

[juanrga]

No, It seems you didn't read the hardware.fr review. It actually proves my prediction and makes your look silly (and completely wrong). Yes, 7900X is tiny bit faster than 1920X when comparing across multiple programs, where some of the programs can't utilize all those threads properly.

They have used lots of programs that scale up above 24 threads, and the average reflects that.

Everytime the programs can fully utilize 24 threads, 1920X is either neck to neck with 7900X or ahead.

That is not true. Examples of workloads that scale above 24 threads and the 12C 1920X is slower than the 10C 7900X because the SKL CPU has much faster performance per core

 

[juanrga]

TR is very impressive for a MCM/NUMA part. The fact that it is in the same league as Intel's monolithic mesh networked Xeon parts is quite remarkable.

The disadvantages of the MCM approach versus monolithic die are two: higher latency and higher power consumption.

Highly-parallel workloads such as rendering and encoding are throughput-like, and it matters little if you run those in a monolithic die or in a four-socket system, because those workloads aren't sensitive to latencies. Blender, V-RAY, Handbrake,... run fine on MCM.

The performance disadvantage of the MCM approach comes in latency sensitive workloads, when one core in one die has to interchange information with a core in another die.

The power consumption is higher, but this issue still has to be investigated. The review mentions that they loose watts somewhere between the wall and the CPU, and they speculate if the processors are taking power outside the measured 12V channel. They also detected that AMD is using again the old trick of clocking cores under the base clocks to maintain power under control: "Moving under the base frequency is however something annoying, even if it is not the first time that we see this behavior at AMD".

 

[Osjur]

They have used lots of programs that scale up above 24 threads, and the average reflects that.

That is not true. Examples of workloads that scale above 24 threads and the 12C 1920X is slower than the 10C 7900X because the SKL CPU has much faster performance per core

Wohoo, you bring couple of cases where you show that 7900X can win against 12C 1920X. I never even argued that it couldn't. My line clearly says that 7900X is a bit faster (3%) than 1920X when using avarages (according to the review you chose) but I guess you didn't bother to read that. But why are you even talking about 1920X? Your original claim was that 1950X will lose against 7900X when using <24 threads. My claim is that it will not on most workloads, until the thread count drops to 14-15 threads and I've done the math.

I do like Hardware.fr testing methodology but the ram (2400mhz) they decided to pair with TR gives it a big perf. penalty.

 

[Gideon]

Wohoo, you bring couple of cases where you show that 7900X can win against 12C 1920X. I never even argued that it couldn't. My line clearly says that 7900X is a bit faster (3%) than 1920X when using avarages (according to the review you chose) but I guess you didn't bother to read that. But why are you even talking about 1920X? Your original claim was that 1950X will lose against 7900X when using <24 threads. My claim is that it will not on most workloads, until the thread count drops to 14-15 threads and I've done the math.

I do like Hardware.fr testing methodology but the ram (2400mhz) they decided to pair with TR gives it a big perf. penalty.

Low speed memory like that will harm the AMD system more then Intel due the the Fabric speed, he knows that and why hes picking that review site.

 

[gigaxtreme1]

Here's a cool article. It's WCCF, but seems like a cool thing to do by the green team. Little off topic but in the genre.
http://wccftech.com/nvidia-welcomes-amds-x86-comeback-endorses-threadripper-amd-high-fives-back/

 

[ecmaster76]

The disadvantages of the MCM approach versus monolithic die are two: higher latency and higher power consumption.

Highly-parallel workloads such as rendering and encoding are throughput-like, and it matters little if you run those in a monolithic die or in a four-socket system, because those workloads aren't sensitive to latencies. Blender, V-RAY, Handbrake,... run fine on MCM.

The performance disadvantage of the MCM approach comes in latency sensitive workloads, when one core in one die has to interchange information with a core in another die.

You are correct but my point is they minimized those disadvantages most impressively and in such a way as to allow Chimpzilla to viably compete the 800lb Gorilla's 800lb HEDT platform

MCM allows them to keep yields higher vs a larger die and saves them from having to produce additional masks and wafer runs for TR and Epyc. In fact they could stockpile unpackaged known good dies and use them to produce AM4, TR4 and SP3 parts on an as needed basis depending on market response.

If they can keep some inertia going I think the situation could shift further in their favor with the next Zen revision. Chipzilla is locked into to Skylake and Kaby Lake HEDT/Xeon designs already and other than more cores it doesn't sound like Coffee Lake is going to change things much. Intel faces either putting up with this for a year or two or meaningfully lowering prices but I doubt they'll do the latter.

 

[OrangeKhrush]

It seems my prediction Intel 10C chip would be faster than TR even under 24 threads workloads was rather good.

Also we see how both TR models are slower than the 1800X on games, as expected. I wait anxious to 1900X vs 1800X comparison.

You mean the 6950X is not a better gaming part than a 7700K?

Sometimes when I think it couldn't get anymore obvious.

If the Spanish football team could move the goalposts like you do, they probably wouldn't have conceded 20 goals in the last world cup.

Firstly this is a general computing part, it caters for a specific user likely doing large content production and it is very good at that, since AVX is very unpopular among developers in mainstream applications due to Intel's iron grip the issue of AVX and the shifting of that goal post is irrelevant to the argument.

Secondly, gaming on HEDT is the worst kind of diminishing returns, it is not like Intel HEDT has ever been an appealing platform for gamers nor an affordable one. So lets not treat this like an anomaly, ever since the 4790K came out HEDT is only prevalent when it came to lanes and prestige but no real gamer will give up a 7700K for a 7980XE similarly nobody will take a 1950X for gaming when a 1700 or 1600 is good enough

 

[OrangeKhrush]

No, you are reading it wrong.

Stock clock scores with my math:
i9 7900X Score: 2200p
i9 7960X Score: 3168p (math to get this score: 2200p * 1.6 * (3.6/4.0)
TR 1950X Score: 2900p (math to get this score: 3400p * (3.5/4.1)

Second math part was for 7900X running against TR 1950X while both are utilizing 24 threads.

Oh well, my math was bit wrong, TR is actually getting little over 3000p in CB and its because I thought the all core turbo would be 3.5Ghz and I calculated that it would only get 2900p with that.

The real all core turbo for TR is 3.7Ghz.

EDIT: pic of the day (from one techsite):

Look how big the difference is

Kyle needs to make his graphs look like this.

Well if you separate one with decimal points it makes it look impressive. I think everyone is laughing

 

[juanrga]

Wohoo, you bring couple of cases where you show that 7900X can win against 12C 1920X. I never even argued that it couldn't. My line clearly says that 7900X is a bit faster (3%) than 1920X when using avarages (according to the review you chose) but I guess you didn't bother to read that. But why are you even talking about 1920X? Your original claim was that 1950X will lose against 7900X when using <24 threads. My claim is that it will not on most workloads, until the thread count drops to 14-15 threads and I've done the math.

Well, you and others objected to my pre-launch claim that, due to higher performance per core, the 10C Skylake "would be faster" than the 16C TR on workloads with 24 threads. I am talking about the 1920X, for two reasons (i) because we don't have workloads with exactly 24 threads running on the 1950X, and (ii) because the equation I used for my claim applies to both the 1950X and the 1920X, because it is a basic equation of computing,

I do like Hardware.fr testing methodology but the ram (2400mhz) they decided to pair with TR gives it a big perf. penalty.

TreadRipper with 3200MHz RAM

When 10C SKL wins to 12C TR, it usually does by a large amount that when it loses. I have not computed the average, but I am sure that it will be close to the average reported by HFR review, despite the above benches are all encoding/rendering, whereas HFR average is for a broader range of applications.

 

[juanrga]

Low speed memory like that will harm the AMD system more then Intel due the the Fabric speed, he knows that and why hes picking that review site.

Just in the above post you can find 3200MHz benches. Overall, the conclusion is the same than from the HFR review: 10C SKL is a bit faster than 12C TR even in throughput-like workloads (encoding/rendering) that scale well above 12C and are favoring the Zen muarch (which is optimized for throughput).

 

[JustReason]

Its ok. No need to point out the 16Cpower usage against the 7900X or state the fact that CPU usage was not fully utilized on TR. Yeah no need to point out that with just a little optimization TR will only get better and based on previous Ryzen optimizations it can be quite huge. And no need to point out the platform advantage TR has with higher IO and thus far better operation as far as power delivery and temperature.

 

[N4CR]

Its ok. No need to point out the 16Cpower usage against the 7900X or state the fact that CPU usage was not fully utilized on TR. Yeah no need to point out that with just a little optimization TR will only get better and based on previous Ryzen optimizations it can be quite huge. And no need to point out the platform advantage TR has with higher IO and thus far better operation as far as power delivery and temperature.

Lol they hate the power usage question, best part is it doesn't matter now Intel is behind all of a sudden. Can you remember when 30W difference was literally like joseph mengele himself stoking the furnaces of auschwitz? The great PCI slot-o-caust comes to mind too.
'waves hands' And ignore the fact that the 1950 and the 7900 are the same price, AMD suckers! We must only focus on 24 threads, it's all you'll ever need.

What I find most hilarious is how close the 1800x is and that a 1700 could easily OC to that and beyond for 1/4 of the price of the 7900 or TR lol.

Oh and before I forget.

ECC for HEDT

 

[juanrga]

Its ok. No need to point out the 16Cpower usage against the 7900X or state the fact that CPU usage was not fully utilized on TR.

With workloads that fully load all cores, the 16C TR is drawing more power than the 7900X as expected, but the interesting part here is that AMD is using a trick on TR to maintain power under control: cores are downclocked under base frequency when the CPU is too loaded: "Moving under the base frequency is however something annoying, even if it is not the first time that we see this behavior at AMD." Reviewers also noted that some watts are missing in the way from the wall to the socket. Their current hypothesis for this discrepancy is that the CPUs are drawing the missed watts outside the ATX12 channel: "Which makes us wonder if these processors would not draw a portion of their power from the 24-pin ATX connector."

 

[ecmaster76]

With workloads that fully load all cores, the 16C TR is drawing more power than the 7900X as expected, but the interesting part here is that AMD is using a trick on TR to maintain power under control: cores are downclocked under base frequency when the CPU is too loaded: "Moving under the base frequency is however something annoying, even if it is not the first time that we see this behavior at AMD." Reviewers also noted that some watts are missing in the way from the wall to the socket. Their current hypothesis for this discrepancy is that the CPUs are drawing the missed watts outside the ATX12 channel: "Which makes us wonder if these processors would not draw a portion of their power from the 24-pin ATX connector."

Is it thermal throttling or power throttling? Either way they have top do something and underclocking is better than crashing

Got any links on the power thing? The 24pin connector has 12v lines to and, well, its there to provide power to all the things including the CPU. Before the Pentium 4, the 20(!) pin MB connector was all we had so there is nothing wrong with using it.

 

[FrgMstr]

With workloads that fully load all cores, the 16C TR is drawing more power than the 7900X as expected, but the interesting part here is that AMD is using a trick on TR to maintain power under control: cores are downclocked under base frequency when the CPU is too loaded: "Moving under the base frequency is however something annoying, even if it is not the first time that we see this behavior at AMD." Reviewers also noted that some watts are missing in the way from the wall to the socket. Their current hypothesis for this discrepancy is that the CPUs are drawing the missed watts outside the ATX12 channel: "Which makes us wonder if these processors would not draw a portion of their power from the 24-pin ATX connector."

The TR does some strange stuff when it starts getting thermally loaded, and I would suggest that NO ONE had really good cooling for it for reviews, expect us.

Hasn't the processor always drawn some of its power from the 24-pin? I think the inclusion of the 8-pin was a supplemental approach. Maybe I am wrong on this. I have learned this, if there are extra 8-pin/4-pin ATX down on the motherboard, you need to use those.