Ryzen vs i7 (Mainstream); Threadripper vs i9 (HEDT); X299 vs X399/TRX40; Xeon vs Epyc

The Threadripper 3950x leak; https://browser.geekbench.com/v4/cpu/search?utf8=✓&q=amd+sharkstooth

Nearest Intel the 28 core w-3175x ($2999); https://browser.geekbench.com/v4/cpu/14240037

 

i wonder what the TR naming scheme will be, since 3950x will already be taken by the 16 core mainstream chip coming this sept

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

Probably, if there is a 16-core, something like a 3955X or similar, then 24-core is 3960X, 32-core as 3970X, 48-core (if they make one) as 3980WX, and 64-core is the new 3990WX would be my prediction.

 

no point if the 16 cores can only hit like 4.2 - 4.3ghz at absurd voltage like 1.4v.

more interested in seeing how new TR silicon will bin, everything else now is secondary.

 

Except silicon lottery binning on the 3900X and 3800X show the top bins do that at the 1.2V-1.3V range. So if these are binned better than mainstream, should be looking at 50-150MHz, as TR also spreads the heat over a much larger surface area.

 

Silicon Lottery Reveals AMD Ryzen 3000 Binning Stats
by Zhiye Liu August 14, 2019

Starting with the AMD Ryzen 9 3900X, Silicon Lottery's data show that only 6% of its samples were able to hit 4.20 GHz on all 12 cores with a voltage of 1.250V. Needless to say, this is a pretty disappointing figure for performance seekers who love overclocking their processors for extra performance. On the bright side, 35% of the tested Ryzen 9 3900X chips could do 4.15 GHz on 1.237V.

Source: Silicon Lottery Twitter

 

Everyone knows you can not get much from overclocking Ryzen 7nm. As far though as performance seekers, well Ryzen is the ticket.

 

You'll need 3900X

 

< Waiting for 32 core TR3 here .

 

If that 32 core does anywhere near 4.4 all cores, then Intel's OC Xeon is dead, along with with equivalent priced 16 and 18 core Intel extreme CPUs.

 

thats why it is garbage silicon. TR will prob a tiny bit better, hopefully 100mhz.

4.4ghz with zen2's IPC, 32-64 cores quad channel memory thats hard to resist.

 

interesting to see how the 3700x is systematically binned worse than the 3800x

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

They need better silicon for 3800X to compete with Intel's 8 cores chips you know (9900K)

 

This might be true if they were priced equally, but they are not. Even then the 3800x has more cores too.

 

id say theyre doing pretty good on that front

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

More cores? But yeah, better binning = Higher prices.

Cinebench - R15 8xCPU Ranking with a Ryzen 7 3700X

Cinebench - R15 8xCPU Ranking with a Ryzen 7 3800X

 

But look at the voltages. For two scores, both 3700X and 3800X at 4.4GHz, you have to use 1.44V on the 3700X but only 1.363V on the 3800X. Nearly 80mV for the same frequency is nothing to snuff at. Now, if they are using even better bins for the 16-core and 32-core and 64-core threadripper parts, it suggests we will see pretty good performance.

This is especially true for a 32-core under $2K that beats Intel's 28-core when OC'd and costs $3K, plus needs a special $1100-1800 MB to run it.

Edit: Also, this isn't a 1:1 comparison because the 3700X used slower mem, but also used Win 10, whereas the 3800X score is proper in Win 7.

https://hwbot.org/submission/4209071_zalera92_cinebench___r15_ryzen_7_3700x_2325_cb
https://hwbot.org/submission/4197706_jumper118_cinebench___r15_ryzen_7_3800x_2325_cb

 

Then you should expect higher scores for 3700X with equal ram and OS

Yeah, as I said the 3800X is higher binned. But 4.4 is near the top anyway for normal good cooling.

 

3700x and 3800x are fine too compared against a 9900k / 9700k, and small game FPS differences are small enough so as to be unnoticeable. No need to spend $500 on the 3900x or $750 on the 3950x unless you want to outperform the HEDT / Xeon CPU's from Intel.

 

That makes no sense! CB15 does BETTER with Win 10, not worse! You equalize ram, going from 3000 to 3600, you get a boost, but if they did it on Win 7, it drops it down. So you CANNOT say with certainty the 3700X would perform better. That is absurd and if you don't know which chips get better scores with which OS, you shouldn't make such statements. Take my 3707 score in windows 7 versus my 3745 score with benchmate in windows 10. Or any of the geekbench scores. Simply put, your conjecture there makes no sense. Plus 80mV is enough for another 50-100MHz, cooling allowing.

And considering the difference on top frequency for TR on 16-cores each gen, expecting it to stay cooler while running with lower voltage makes sense, which can allow for better speeds.

Either way, we'll see once released and bench scores come out. Same with Cascade-X (which shares the same socket as Skylake-X). Something tells me people will be chaffed at the premium for Intel at same core count, or they will just grab the 32-core chips from AMD at the same price as Intel's 16 or 18 core chips.

 

I haven’t AMD chips so I can’t say how much the OS will do for that chips. But faster ram should help push scores in right direction.

 

Well now I'm informing you of that. GB3, GB4, CB15 are among the ones that, with Ryzen and TR, perform BETTER on Win 10 instead of Win 7. That's why my 4.25GHz 1950X is right behind Mr. Fox's 7960X in GB3 multi while he is at 5.2GHz.




And for GB4:



Now Intel will certainly still get ST top score in GB, but with an extra 15-35% (yes, some Epyc CPUs are seeing a lift as high as 35% on multi-threaded workloads gen to gen), what do you think is going to happen here, especially since in those workloads my heavy water cooling loop is keeping up with a CHILLED 7960X. Further, looking at the scores on Ryzen mainstream, the CB scores seem like TR will be doing very well against these 14nm Intel chips as well.

Then, you tell people they could buy a 32-core chip for the cost of Intel's 16 and 18 core chips and dominate those workloads, the HEDT battle is over.

Also, binning is there and is working. Just wish I could see what those chips could do with my excessive water loop!

Edit: Here are my linux, win 10, and win 7 top scores submitted to geekbench with links:
GB3
Linux - 77106 https://browser.geekbench.com/geekbench3/8675016
Win 10 - 73784 https://browser.geekbench.com/geekbench3/8793286 ( I have the higher score above because done in debug mode and didn't submit the saved file yet)
Win 7 - 64687 https://browser.geekbench.com/geekbench3/8794784

GB4
Linux - 60289 https://browser.geekbench.com/v4/cpu/9715694
Win 10 - 50973 (see above)
Win 7 - 35874 https://browser.geekbench.com/v4/cpu/8726973

There are quite a few benches that do better on Win 10 than Win 7 on Ryzen, plus there isn't the penalty for security patches that windows makes users install on windows 10 for Intel chips, meaning that they don't get dinged as hard, which may be why your spreadsheet has so much working better on Intel chips for Win 7.

 

On another note, people probably remember Intel investing $1B to expand the 14nm node fab capacity.

Well, due to demand, TSMC is investing $6.5B more this year to expand capacity. They cited 5G tech and increasing capacity at 5nm (volume next year) as the reasons for the increased spending.
https://www.digitimes.com/news/a20190814PD207.html

 

at that sort of cancerous voltage with low frequency, we will need to wait a few years before jumping into AMD and AMD laptops.

luckily zen 3 in a yr might see another decent improvement in both frequency & IPC and intel wont have anything "excellent" to be worthy of buying until maybe beginning of 2021 or even 2022. also recently found out ICL ipc is only around 7-8% over CFL and not the claimed 15-18% from intel, at least in CB15.

this guy did a good job revealing intel's issue assuming most of it is true:

Spoiler

I've been saying it for over a year, Intel's 10nm wound up as a tangled up experiment. In short, Intel management wanted such a massive shrink that would ensure their domination of the process market and attract customers to their fabs. The load combined with budget cuts and layoffs (and their other stunts) in the manufacturing group killed 10nm.
In Long: There were three primary mistakes Intel made, Cobalt Metal Layers, Contact Over Active Gate (COAG), and plain ole hubris.

In 2013, Intel was late. The Self Aligned Dual Patterning (SADP) required by the feature size of 14nm had a bad learning curve, yields were very bad at first, to the point where Broadwell was mostly a paper launch in 2014, two quarters behind schedule. This was not a critical problem and it was fixed gradually, such that Skylake was not delayed. Behind the scenes though, the long ramp time created a problem. As Intel has only a single (large) process development team, not leapfrogging teams, the 14nm delay led to a delay of 10nm. The specifications it would shoot for were not set in stone until 2014.

Management gave them a difficult task. To win mobile they had to be power efficient and dense. To win desktop they needed to be fast. To win servers they needed excellent yields. And above all they needed to be better than the competition to attract new customers.
In order to reach the goals set by management, the manufacturing group had to get creative. To that end a number of techniques never put into a production process before were adopted. COAG, SAQP, Cobalt, Ruthenium Liners, Tungsten contacts, single dummy gate, etc. This push is directly what led to the death of the process. Of those, only really COAG and Cobalt are causing the issues. I'll go into the specific problems next.

If anyone is to blame, its the management, and their firing of the CEO with a ******** reason shows the board will not accept responsibility for the companies failings. They will not come clean in the foreseeable future. Their foundries are virtually dead after all the firings and cost cutting.

The idea with Contact Over Active Gate is that instead of extending a gate such that it connects up with a contact to the side (thus using space on the side), the Contact stretches directly from the metal layer to the gate, rather than laying onto the substrate. This means there is NO room for error on manufacturing. The slightest misalignment leads to ****ed contacts. Thermal expansion, micro-vibrations from people walking nearby, changes in air pressure, imagine a cause, and it'll affect yields. I bet you the bloody position of the Moon can affect it. This kills the yields.

To hit the targets Intel set, a minimum metal pitch of 36nm was selected. When you have Copper wires on a process they need to have a liner around them, this prevents diffusion, electromigration, and other nasty electrical fun. But this liner needs to be a certain thickness, so when the overall size of the wire gets smaller, the liner takes up a larger portion of it. Below 40nm it was thought that Cobalt would have superior electrical properties, despite it having a higher bulk electrical resistance. Its far more resistant to electromigration and needs a miniscule barrier to prevent it, while its resistance decreases at a slower rate as the wire size gets smaller.

However, Intel overlooked two key problems: ductility/malleability, and thermal conductivity. Even at those tiny levels, Copper wires would be able to handle thermal expansion mechanical loads, bending and stretching ever so slightly as a processor made its rounds. And copper is Very good at transferring heat, letting the lower metal layers sink heat into the upper ones. Meanwhile Cobalt is hilariously brittle and has a sixth the thermal conductivity as Copper. On operation hot spots start to form, heat can't get away, brittle nature creates microfractures, and higher voltage to cross the fracture boundaries. Means the voltage/frequency curve is hilariously bad. This kills the performance and power usage.

So where does it leave us at?

10nm was meant to launch end of 2015, after 14nm this was pushed to 2016. It was Q3 2018 when i originally wrote much of what is outlined here, and the only 10nm chip at that time was a minuscule dual core made in a one-off batch of 100k units that took 6 months to assemble. Yields are sub 1%, the GPU doesn't function, and power usage was higher than 22nm.

TSMC and at the time GloFlo were both ramping their 7nm processes, and while they're comparable to the 10nm in density, they actually work. They both are using SAQP for the transistors, but the choice of 40nm metal pitch allowed for SADP for that particular layer. GloFlo was using Copper with a Cobalt/Tantalum liner while TSMC is using straight Copper/Tantalum. Neither screwed with COAG or Single Dummy, but GloFlo was using Cobalt contacts.

TSMC 7nm has been volume production, GloFlo meant enter it in a few months after TSMC initially did, but has since put it on hold indefinitely in favor of pursuing nodes beyond that. Consumer products on both of them meant to enter the market in mid-2019. TSMC did, latter didn't due to aforementioned reasons. Regardless, They're both going to outperform Intel's 10nm. Their manufacturing group failed at all levels. To make matters worse they didn't even bother backporting their improved cores to 14nm. Icelake has been inhouse since early 2017. The design is finalized, they just can't make the damn thing and didn't backport it. Management thought they couldn't fail.

It is in my personal opinion that their 10nm process will never be financially viable. They were literally creating 10nm micro housefires that burned their own chips.
Cannonlake is virtually dead. Ice Lake-U and Y (sub 15w mobile dual cores) came out as predicted still in mid-2019. I doubt Ice Lake-S (quads) will see anything beyond a paper launch ala Broadwell quads. Ice Lake-SP is a dead proposition, there's no way they're going to be able to make a server core based off that 10nm. Tiger Lake and Sapphire Rapids? Who the hell knows. Why worry about something that isn't going to happen.

Intel's chance lies with 7nm, but people already predicted it won't arrive until 2021 at the earliest. After Cascade Lake is Cooper Lake; Kaby Lake-U is being replaced by Whiskey Lake, which'll be replaced by Comet Lake to run alongside the Ice Lake-U/Y dual cores. Amber Lake will replace Kaby Lake-Y until Ice Lake-Y. All those new code names are 14nm, and all are just tweaked Skylake.

Also, The whole 14nm+++++ tripe is meaningless and has been for years. They're standard PDK updates. Everyone in the industry does them. It wasn't until recently that they started being branded, in this case by Intel.

 

So, since the TR 32-core GB4 score was leaked, we now get to see the Intel 18-core Cascade-X GB4 score, which should be priced to compete against each other:


https://www.tomshardware.co.uk/intel-18-core-cascade-lake-x-cpu,news-61453.html
It is "3.36% [faster] in single-core workloads and around 7.4% [faster] in multi-core workloads" compared to the 9980XE.

Compare this to the 32-core below, priced the same, approximately:


"The Castle Peak chip performs up to 4.72% and 14.63% faster than the Threadripper 2990WX in single-core and multi-core workloads, respectively." This is likely while also not seeing the prior gen's performance regression due to using the I/O chip.

Looks like faster single thread at stock, plus a massive 74% faster at multi-threaded workloads. If I was Shania Twain, I'd have to say of Intel "that don't impress me much."

Overclocking is yet to be seen on the Cascade-X chips, but the Skylake-X originally, for the 7 series chips, used 14nm+ (Kaby) process. I do not know if the refresh (9 series) used 14nm+ or ++ variant. If not ++, there might be a little more in the OC tank, but otherwise, not really.

Meanwhile, what was pointed out by a person in another forum:
"Looks like the first sample was running all 64 cores between 4150 - 4200 Mhz. The second sample logged clocks between 4150 - 4350 Mhz.
https://browser.geekbench.com/v4/cpu/14281644.gb4
https://browser.geekbench.com/v4/cpu/14281648.gb4 "

 

If all used Linux as the OS I would be more impressed by the charts. I did use the browser and compared the all Linux scores and the 32 core AMD easily beat out even Intel's 28 core in geekbench.

 

Intel Tiger Lake-U 4 Core, 8 Thread Processors With Gen 12, Xe GPUs Spotted – Faster Than Core i7-8700K In Single-Core Tests at Just 3.6 GHz Clock Wccftech.com
Now, two Tiger Lake CPU entries have appeared on the user benchmark database along with early performance numbers. Both CPUs are part of the Tiger Lake-U family which includes 15-28W SKUs. Both processors are configured as a 4 core and 8 thread design, featuring clock speeds of 1.2 GHz base and 3.6 GHz boost. Since the launch is a year away, these clock speeds don’t look final & may end up being much higher in the final retail variants.

 

thats +30% IPC over CFL-R and still 23% over ryzen 3000

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

How do you get 30%? On quad core, I get around 26.6%. [Edit: nevermind, whereas AMD gets a higher percent jump with multicore, Intel has a higher percent on single core.]

Also, on pure performance on quad core score, it is still lagging Zen 2, while this uarch is made to go against Zen 3, which is already finished and ready for tapeout. The 3700U and 3750H are Zen+ while the mobile chips it will go against are Zen 2 cores, which is why I selected the 4-core score and looked at the 3900X. If the mobile parts next year can reach anywhere near the 4.3GHz range like most Zen 2 chips do, then that is a solid measure for comparison.

Also, Ice Lake -SP comes in Q2 2020. Tiger Lake - U comes around Q2-Q3 currently, although it may be pushed out later. Desktop gets 14nm Comet Lake-S in Q4 to Q1 2020 and Rocket Lake-S comes around a year later for desktop, also on 14nm. That means, since these numbers are around what current Zen 2 gets, even with a speed boost, it will have a hard time against Zen 3.

Kind of sad to see the peak.

Edit: @Papusan - didn't you say it was pathetic AMD couldn't hit 5GHz, and that the speeds were bad. Yet now you herald even slower speeds as magnificent. Why the change in position?

 

I still mean low clocks is pathetic (even more pathetic if all cores can't be put equal max single core boost). Where did I herald my own meaning that even slower speeds as magnificent in mentioned post #2927? I just post info found from the web. I have two choices, bruh. Either not post... Or post

 

Fair enough. Wanted to make sure it was just sharing info and not trying to make claims on Intel's performance there.

Intel's 10nm kinda sucks, especially on frequency (the architecture is great! Imagine that uarch on TSMC 7nm! May not be the frequencies you want, but that would likely DECIMATE AMD's Zen 2 chips if the frequency jumped up to anything near AMD's current clocks).

 

You or/and others should know where I stand. High IPC is well and good, but High clock speed is needed to get out most possible performance. Higher the better. One of this two ain't good enough.

 

Just like you know my position. I don't care how it gets there, just that it gets there. Enough IPC and frequency doesn't matter. Just raw performance. As to what I buy, there is also a value component to that performance.

But I hear ya.

 

i think the value component applies to all of us. some just value performance more and willing to pay a bit more. right now its hard to justify buying intel really, only very handful of scenario people would go for intel.

tiger lake ~20-21% IPC over CFL
https://wccftech.com/intel-tiger-lake-u-cpus-gen-12-xe-gpus-spotted-tested/

remember we waited 8 generation for 20-25% IPC improvement from sandy to skylake, now in 2 yrs we get 20% cause of AMD. can't wait to see zen 3 or 4 versus icelake/tigerlake.

 

I thought everyone agreed User Benchmark is worthless.

 

oh that was just clock for clock, i didnt consider core count since were talking single thread here, anyways....

well as long as its same number of cores between two samples i guess one might still use them

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

Agreement isn't required for it to be worthless. That stands on its own regardless of what some might think. But, for the record, I do agree.

 

Worthless test or not... Higher scores still beat the lower one If it's correct or not (the User Benchmark)... Depends probably what brand you prefer. See... Lower can be better

 

Ryzen 7 3800X Show Different Boost Clocks with different motherboards Guru3d.com | 08/22/2019

There has been a theory for a while not that motherboard manufacturers tweak their products a bit to squeeze out the best performance of AMD processors. Basically, people have been seeing small performance differences and some have complained that the advertised boost frequency is not even reached.


Cinebench R20 was run three times in a single-core test and that will read out the maximum boost clock monitored with HWiNFO64. The AMD for the Ryzen 7 3800X reached a boost clock of tan advertised 4,500 MHz on a number of six motherboards. Then three mainboards hit 4,475 MHz and two 4,465 MHz. A total of three mainboards could not reach a higher boost clock than 4.375 MHz

 

For such newly released CPU's and motherboards, that's pretty good that only 3 out of 15 failed to boost to the full capabilities of a particular sample CPU. Hopefully firmware updates will bring those 3 up to spec soon too.

4.5ghz boost for the 3800x, 4.6ghz boost for the 3900x, 4.7ghz boost for the 3950x, I wonder what the ThreadRipper's will boost to?

Maybe sooner than later next year we will see Ryzen boosting to 5ghz? At this point the IPC advantages of Ryzen over Intel allow the Ryzen CPU's to outperform Intel in single core even with Ryzen not boosting as high as Intel. When Ryzen boosts the same or higher than Intel it's all over for Intel --- for a 3rd time.

 

we shouldnt have to worry about 25-75mhz.. the only time we should really worry about it is if temperature or voltage is too high for the additional 100mhz. can't be satisfied until we see something close to 5ghz range. 4.8 is a good start, for all cores frequency.

 

Yup, but it's nice to see it fall on the plus side than fail to make the specification posted - even if it is such a small unnoticeable amount.

If we didn't pay attention to such details then the vendors won't either, and slowly slip further from making the specs as advertised.

It's good to check on them to "keep them honest".

 

Yeah, I'm more interested in the rumored two different chipsets for TR, one of which may have 8 memory channels. If true, I may have to wait for next year to upgrade and grab that (and a new MB)!

That would be the upgrade that I want!

 

If I paid for 5.0GHz I wouldn’t be happy if I can’t get it. Don’t forgot 75MHz is nearly 1 bin bruh. Or near 40 point in CBR-15 with a 8 cores chips

 

I thought the 7700k / 8700k / 9900k wouldn't run all core 5ghz in your laptop? Didn't you "pay for 5ghz" then too? Realistically your laptop won't run 5ghz on all cores under a constant 100% load without becoming unlivable - it's cooling fans are going to be way too loud.

Every motherboard / laptop is different, that's why we watch the reviews to find out what works well together and gets the best performance for cost - the best value for money or the highest performance - based on your personal priority.

All of those motherboards are close enough, with 3 of them as outliers that still need tuning to catch up with the group. It's good to know ahead of time what's what, and if I had one of those motherboards I'd do the testing myself and send in a complaint to their support to get it fixed.

If I liked the motherboard otherwise, I don't think I'd return it only based on that shortfall. There'd need to be a number of problems to make me disassemble that whole PC - return that motherboard - get another motherboard and put it all back together.

Being forewarned though - knowing before the build - I'd pick another one of the top performing boards - but tempered by cost and other features. I don't need to pay out $100's more for a slightly measurable and likely unnoticeable improvement.

 

The only chips I couldn't come near 5.0GHz outside Cpu-Z validation was the 6700K. This is 3800x class...
http://forum.notebookreview.com/thr...ers-welcome-too.810490/page-640#post-10943952

 

Which one do you have running now? Do you run it all the time at 100% all core 5.0 ghz? "Reaching 5.0ghz" isn't running it 24/7 as your daily driver tune.

I could have sworn you often posted sub 5.0ghz results... sorry if I am mis-remembering. Nice to see your 9900k reached 5.0ghz yesterday.

 

5.0GHz all cores 24/7 is too much for the old lady I need the TM board for that. But stock with an undervolt is ok.

 

thats a problem with the laptop though too small of a heatsink. if our chip pair up with gt76 titan or alienware heatsink, it has the capacity to run 5.1ghz 24/7.

 

With my 6700K, I was at 4.8GHz all day, but that was a desktop with a 360mm AIO. Benching was 5.0-5.1.

Meanwhile, seems the Ryzen, even with that lower frequency, is better. Now, if going with gaming, the 7700K is similar in scores to a 3600. I think 3DMark still favors Intel a good bit.

Even with that, I'm more about the multi, not the single thread scores. That favors AMD at any given price point.

But considering now we see that different bios implementations are effecting the boost algo, that should be, hopefully, resolved by TR, which will have higher boost clocks. So things really can get interesting. Also, there is a chance of the 3950X beating the 7960X and 9960X, which would mean an inflection point on per core values, something you care about. We will find out in a couple weeks on that. For sure the TR will though, so.

Edit: caveat on meaning air and water. Chilled and LN2 may differ on which is better.

Edit 2: The single core boost on Cascade may change that calculation, don't know though. Very little out on it.

Also, Intel still excels at Prime calculations, depending on the program, it seems.