Jump to content
3D Coat Forums
Sign in to follow this  
Carlosan

Battle of the Cores: Intel’s Core i9 vs AMD’s Threadripper vs...

Recommended Posts

https://wccftech.com/goldman-downgrades-intel-while-upgrading-outlook-for-amd/

 

Goldman Sachs not only upgraded AMD (NASDAQ:AMD 19.06 0%) from “sell” to “neutral” today, but Analyst Toshya Hari cut the rating on Intel (NASDAQ:INTC 48.85 -0.03%) from “neutral” to “sell,” doing a complete polar swap on the CPU giants. In Hari’s note to clients, he said “[w]e find it increasingly harder to argue our prior bear thesis — even following the recent stock price move — given Intel’s struggles with 10nm process technology.”

This after my previous article regarding Morgan Stanley saying the general semiconductor industry is on the downswing, in addition to our article regarding Intel’s 10nm process is vastly superior to any current 10nm process currently in mass production as well as being a more dense and versatile package, it is still an unknown factor whether or not Intel can deliver, but the same thing could be said of AMD. While some analysts believe the smaller the process, the better the chip, not all processes are created equal.

 

From April 6, 2017, when Goldman downgraded AMD to sell, AMD stock has gained 44% versus the S&P’s total return of 21% over the same period of time. With AMD pushing hard for market gains in the CPU market, it’s clear analysts are noticing something different, and not necessarily just Intel’s recent and scandalous change in leadership.

Though this also could have analyst worried. Brian Krzanich, who resigned after having an affair with a subordinate, has not formally been replaced. Bob Swan was named interim CEO as of June 21st, but Swan was the CFO prior to all of this and made comments during an all-hands staff meeting that not only was Krzanich asked to resign for breaching company policy in regard to having a relationship with a fellow employee, he was removing himself from candidacy for CEO, further questioning the next person in charge. He also made clear their company’s sexual harassment policy.

Now that Goldman has changed its tune on AMD, it also had quite a bit more in regard to it’s now also changing their opinion of Intel. “We believe Intel’s struggles will provide a reason for customers (PC OEMs, server OEMs, Cloud Service Providers, etc.) to adopt, or at least investigate, AMD’s product.”

 

Their now adapted appraisal of 2019 lines out a sizable gain of 5.1%, moving into 2020 where they could see 9.4% gains in the server chip market alone. The desktop and OEM markets are different stories entirely, but if similar gains are seen here as well, Goldman could continue down this path. The playing field, from Goldman’s perspective, seems to be leveling off a bit.

The next two years in the CPU market could be huge, not only for competition in the stock market but for the innovation it will provide everyone. This is just starting to get interesting. Affairs, reversals of ratings and even Goldman Sachs admitting they were wrong about a previous prediction. This story has got it all, folks. Stay tuned, and we’ll keep you informed!

 

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites

I was going to run out in September and buy the 2950X and even the latest NVidia GPU, the 2080 or whatever it's going to be called, but my friend an EE reminded me of Black Friday.  Something to remember from last year..

https://www.forbes.com/sites/antonyleather/2017/11/23/amd-and-intel-in-massive-black-friday-processor-price-war/#6147e22d770a

Threadripper 1950X: Launch price $999 - Average price $949 - Black Friday Sale price $799 (saving $150) a 16% saving.

A similar 16% reduction in price on the AMD Threadripper 299WX 32 core CPU would be $287 USD.

Oh well I wait a couple extra months....but by then they'll be letting out leaks on the soon to appear 64 core 128 thread Ryzen CPU..

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites
On 8/11/2018 at 3:49 PM, L'Ancien Regime said:

Sort of a bit off topic but since we're thinking about next builds, how about a 1TB SSD for $173.00 USD?

 

https://wccftech.com/review/crucial-mx-500-1tb-ssd-review-one-of-the-best-in-its-class/

That's a good find, but I think it's time for me to finally go M.2. Been wanting one for a while, but the performance and price of SSD's have been "good enough" to put it off for a while. These are definitely still good choices for media disks, when using After Effects and Premiere Pro, etc.

Share this post


Link to post
Share on other sites
3 hours ago, AbnRanger said:

That's a good find, but I think it's time for me to finally go M.2. Been wanting one for a while, but the performance and price of SSD's have been "good enough" to put it off for a while. These are definitely still good choices for media disks, when using After Effects and Premiere Pro, etc.

Yeah I've been looking at those for a year now and that's definitely lthe way to go. That SSD would be good for a RAID array.

But I remember paying $500 for 500 gb SSD  just a few years ago..

intel.web.416.234-100613940-orig.jpg

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites

That Ryzen 2950X (2nd Gen of the 16-core 1950X ThreadRipper CPU) is looking like a great value for the average freelancer and small studio. I'd love to see how much it helps 3DCoat. Probably most noticeable in the Paint Workspace, working with 8k maps or larger. 3DCoat can be a memory hog when working large assets. I was doing UV cleanup of the (Felippe Beckman's) Viking model and when I went to APPLY UV, it was soaking up 47GB!!!

Andrew has to do something about that and sliders in the Paint Workspace. One of his developers is working on a GPU brush engine for the Paint Workspace and he said it is supposed to use less memory, so we will have to see.

Share this post


Link to post
Share on other sites

I'm seriously reassessing my purchase of of the 2990WX in light of NVIDIA's  CEO Jensen Huang's keynote speech at Siggraph 2018 in Vancouver last night. 

 

The purpose from my POV was always render buckets. 64 render buckets in Mantra or Vray etc sounded like the solution to my main bottleneck, which is in creating and testing new textures in full ray trace mode.

The announcement last night is that with RTX, real time ray tracing had arrived at an affordable price (for me).

I'm not sure yet though. It's going to be interesting to hear all the discussions on this development over the next month or two.

 

You can watch it here;

 

https://wccftech.com/watch-nvidia-ceo-siggraph-2018-livestream-here/

 

All the details and analysis here;

https://wccftech.com/nvidia-geforce-rtx-graphics-card-turing-launch/

https://www.cnet.com/news/nvidia-turing-architecture-arrives-in-high-end-quadro-rtx-gpus/

rtx.JPG.5c4b9d799473ed312eb9420b33241190.JPG

 

If $10k sounds like a lot it is but it's actually a bargain compared to this

 

1164945012_nvidia1.thumb.JPG.1397ce428e63851270a78fed89d2fc3b.JPG

 

The $10000 top of the line Quadro RTX is the equivalent to this DGX worstation for $50k. (Reminds me of the days of SGI workstations)

I could see buying the Quadro RTX 5000 for $2300 USD

Rumors are the Turing Nvidia RTX 2080 is going to be $3000 or so the rumors say.

 

Watch the video of Huang's speech closely. It's full of pretty impressive stuff.

 

Maybe the answer is a new second generation Ryzen 12 core that's a lot cheaper, with 128 gigs of DDR4 RAM and one of these Turing GTX GPUs for real time ray tracing. At the end of th video Huang says Maya et al are already on board with this so it'll run on all the render engines no problem. 4k, even 8k real time ray traced renders.

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites

How is RTX going to render scenes with engines that are CPU-based? A 32-core CPU will come in really handy when a GPU engine isn't available or you get Out of Core issues. Not to mention how much it can help with simulations. The GPUi

Share this post


Link to post
Share on other sites

...the GPU is taking a lot of the CPU's workload, but there is still a lot of CPU-based calculations needed, everywhere you turn. So, if you already bought the hardware, I'd stick with it.

Share this post


Link to post
Share on other sites
1 minute ago, AbnRanger said:

How is RTX going to render scenes with engines that are CPU-based? A 32-core CPU will come in really handy when a GPU engine isn't available or you get Out of Core issues. Not to mention how much it can help with simulations. The GPUi

That's exactly what I was asking last night as I watched the video.  Watch the video carefully to the end. It's packed with very interesting information that may revolutionize  your opinions on what to buy next.

 

I"m not sure now. Your questions were already foremost in my mind but now things have changed.

 

Also I noticed that even thought they're claiming it does real time SSS there were no good examples of it in the video though I may have missed that part as I prepared my dinner..

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites

See? Industry is already ahead of the curve on this, working in concert with Nvidia

 

http://www.cgchannel.com/2018/08/video-chaos-groups-neat-new-real-time-ray-tracing-tech/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+cgchannel%2FnHpU+(CG+Channel+-+Entertainment+Production+Art)

 

Chaos Group has posted a demo of Project Lavina, its new real-time ray tracing system based on Microsoft’s DirectX Raytracing (DXR) technology.

The video shows a standard .vrscene file with a total of 300 billion triangles being rendered at over 25fps on a system containing a single Nvidia Quadro RTX 6000 GPU.

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites

The Quadro options are pretty pricey but this price list for the Nvidia GeForce RTX 2080 coming this week show a more affordable price range. Same GPU as the Quadros but faster (for gaming) and perhaps less  CUDA cores than Quadro series.

 

gtx.thumb.JPG.a27be788c4961d270f27046af14f26c0.JPG

 

At $699 I could get 2 of those and really boost my power. Threre's a special new NVlink for Nvidia Turings that make them operate effectively as one.

Also if you watch that keynote speech the CEO is saying that these GPUs will handle the heavy lifting on photostitching and 3d photogrammetry jobs too.

2844843-a.jpg.be7f516a03fc33fd7946c683b193f599.jpg

Turing-GPU-1-1030x515.thumb.png.6eaee9ba2ca07852bd18d5116b7ca1a1.png

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites

Previously, when NVIDIA announced their RTX real-time ray tracking technology, there were questions about whether or not the technology would get any significant traction among developers. At the SIGGRAPH 2018 press conference, NVIDIA’s CEO Jensen Huang announced a slew of professional ISVs that are adopting NVIDIA’s RTX technology to enable its use in their software. These include programs like Adobe Dimension CC, Autodesk Arnold, Clarisse, DaVinci Resolve, Dassault Systemes Catia and Solidworks, Octane Render, ParaView, Redshift, Siemens NX, Unity and Unreal Engine. Having all of these companies onboard with RTX means that there’s a much higher probability that RTX will gain enough traction in the industry.  Principal Analyst Patrick Moorhead summed up ISV support saying it was a "done deal".

https://www.forbes.com/sites/moorinsights/2018/08/14/nvidia-doubles-down-on-ray-tracing-with-turing/#494606665bb2

 

Amazing to see Clarisse included in there; I've talked to Sam Assadian, its founder and he was pretty adamant that it would always be  CPU intensive and they wouldn't bother to go the GPU route..

 

In addition to giving details on the Quadro RTX series and the Turing GPU, NVIDIA also announced yet another appliance reference architecture. This new reference architecture from NVIDIA is called the RTX Serve and is designed to serve the needs of VFX studios and other companies doing visualization wanting photorealistic rendering with ray tracing in real-time or near real-time. Each NVIDIA RTX Server features eight RTX 8000 GPUs inside at a price of $500,000 which according to NVIDIA is a steal compared to $2 million worth of CPUs that it would take to accomplish the same amount of rendering. These servers are designed to sit rack-mounted in a data center or a cabinet so that the users can have quick access to a lot of rendering horsepower without having to see or hear it.

rtxserver.thumb.JPG.9d054b76b66851cb89f4a7e875f71b76.JPG

Share this post


Link to post
Share on other sites

AMD's 32-core Threadripper dares Intel to match it on cores and price

Whether it's 32 cores or 12 cores, it's all priced to kill Intel.

https://www.pcworld.com/article/3295003/components-processors/amd-2nd-gen-32-core-ryzen-threadripper-2-price-specs-features.html 

 

4cd742df6a609039c64e0c3ad0e33eec4933a586_1_366x500.png

Share this post


Link to post
Share on other sites
3 hours ago, Carlosan said:


The 64 core 128 thread AMD Epyc server CPU is coming out in January 2019.  How long after that do we have to wait until they do some tweaks on it so they can sell it as a 64 core 128 thread  AMD  Threadripper?  

 

If their past strategy in the CPU market are any indication that would mean an AMD Threadripper with 64 cores and 128 threads by August 2020.

 

https://www.servethehome.com/amd-epyc-rome-details-trickle-out-64-cores-128-threads-per-socket/

 

The next generation of AMD EPYC 7000 series is shaping up to be a Xeon killer. The next-generation AMD EPYC 7000 series is codenamed “Rome” and it is going to be a big deal. Instead of adopting Zen+ like the desktop Ryzen CPUs, the new EPYC generation will use the Zen 2 architecture which means improved IPC gains from two generations of core tweaks. Beyond the IPC gains, the next generation parts will be based on 7nm production.

The impact of leapfrogging Intel and using 7nm is several-fold. First, Rome will have up to 64 cores and 128 threads in a single socket. (Edit June 6, 2018: Mea Culpa. Looks like we got some generational information “confirmed” to us incorrectly. Expect a 48 core / 96 thread generation before a 64 core / 128 thread generation. Still quite a huge gap. DDR4 and interconnect improvement information held up to further confirmations. 64 core / 128 thread apparently is still coming, just missed one generation due to a few words not being typed in messages to us.) The new CPUs will be socket compatible with the current SP3 socket motherboards with a small caveat. At STH, we expect Rome to adopt PCIe Gen 4 so motherboards will have to support the higher signaling rates to achieve PCIe Gen 4. We also expect the next generation to have greatly improved Infinity Fabric, an area that the first generation product has room to improve upon.

The other key disclosure is that AMD already has silicon in their labs with the next generation AMD EPYC Rome CPUs in their labs. They will be sampling to partners in the second half of 2018 and will launch in 2019.

This is going to put a lot of pressure on Intel Xeon as Cascade Lake is not going to come anywhere close to the core count of AMD EPYC’s next generation. Intel is scrambling to build a competitive response. 2019 is going to be extremely interesting in the server market.

 

 

Yeah, Intel is having some big problems mastering the 7nm process that AMD has mastered.

https://www.tomshardware.com/news/intel-cpu-10nm-earnings-amd,36967.html

As we pointed out earlier this year, the delay may seem a minor matter, but Intel has sold processors based on the underlying Skylake microarchitecture since 2015, and it's been stuck at the 14nm process since 2014. That means Intel is on the fourth (or fifth) iteration of the same process, which has hampered its ability to bring new microarchitectures to market. That doesn't bode well for a company that regularly claims its process node technology is three years ahead of its competitors.
 

https://www.pcgamesn.com/intel-amd-7nm-cpu-euv

 

But there may be another technology, with its own troubled history, that could finally be close to saving the day for this ageing law: Extreme Ultraviolet Lithography, or EUV.

EUV is a revolutionary new production process that will allow 7nm CPU production to offer higher yields, with lower complexity, and potentially lower costs too. It’s been the holy grail of chip manufacturers for years and is about to become a genuine reality.

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites

I think it would be smart money to wait until the 7nm chips come along. That should help AMD finally reach some serious clock levels. Right now, they don't have any that can comfortably reach 4.5Ghz, despite the high core count. I kind of like that 2950X, though. It seems to be pretty close to the 4.5 mark, with 16 cores/32 threads.

Share this post


Link to post
Share on other sites
5 hours ago, AbnRanger said:

I think it would be smart money to wait until the 7nm chips come along. That should help AMD finally reach some serious clock levels. Right now, they don't have any that can comfortably reach 4.5Ghz, despite the high core count. I kind of like that 2950X, though. It seems to be pretty close to the 4.5 mark, with 16 cores/32 threads.

The same motherboard that the 2950x uses will take the 64core 128 thread Threadripper or so AMD is saying so an upgrade in a year and a half to the 7nm would be available then.

 

I see what you mean on the price point...

 

amd.JPG.26509f538c42624d949e9f1372b58197.JPG

 

 

Edited by L'Ancien Regime

Share this post


Link to post
Share on other sites
On 8/18/2018 at 11:11 AM, AbnRanger said:

I think it would be smart money to wait until the 7nm chips come along. That should help AMD finally reach some serious clock levels. Right now, they don't have any that can comfortably reach 4.5Ghz, despite the high core count. I kind of like that 2950X, though. It seems to be pretty close to the 4.5 mark, with 16 cores/32 threads.

https://wccftech.com/amd-ryzen-threadripper-2950x-16-core-cpu-899-usd-launch/

 

Black Friday and Cyber Monday aren't that far away now.

Edited by L'Ancien Regime
  • Like 1

Share this post


Link to post
Share on other sites

https://wccftech.com/review/gigabyte-x399-aorus-xtreme-motherboard-review/

 

AMD-Ryzen-Threadripper-2000_6-1030x579.thumb.jpg.d3150467bc8e09f5dc4a8465623d9b42.jpg

12 nm LP process technology 
– 1st generation Ryzen and 1st generation Threadripper were manufactured using 14L LPP (Low Power Plus) process technology of GLOBALFOUNDRIES, whereas 2nd generation Ryzen Threadripper based on Zen + microarchitecture was manufactured by GLOBALFOUNDRIES 12 nm LP (Leading Performance) process technology adopted. If the same power consumption is higher than the first generation Threadripper, AMD is appealing that it can realize lower power consumption than the first generation Threadripper for the same clock.


Precision Boost 2 
– The automatic clock-up technology “Precision Boost” adopted by the 1st generation Ryzen and the 1st generation Threadripper had the operation clock determined by the number of loaded cores, but this time the CPU voltage, current, core It has been redesigned to detect the temperature and select an appropriate operation clock. As a result, regardless of the number of cores under load, clock up according to the situation.


XFR 2 (Extended Frequency Range 2) 
~ “If the CPU temperature condition permits,” XFR “to operate with a higher clock beyond the maximum clock of Precision Boost becomes the 2nd generation, and as with Precision Boost 2, the number of cores is restricted lost. Depending on the performance of the CPU cooling system, the performance will improve up to 7%


Reduction of access delay of cache and main memory

– Access delay to cache and main memory is smaller for first-generation Threadripper. Up to 13% improvement in L1, up to 34% in L2, up to 16% in L3, up to 11% in main memory, resulting in a 3% increase in instruction execution count (IPC, Instruction per Clock) per clock It is said thatDSC_0698-Custom-1030x687.thumb.jpg.0f982d0faad4037ad14eeec6cd4425cd.jpg

  • Like 1

Share this post


Link to post
Share on other sites

https://wccftech.com/amd-epyc-rome-7nm-64-core-cpu-performance-benchmark-leak/

 

Alleged AMD EPYC ‘Rome’ 7nm Based 64 Core Processor Performance Leaks Out – Scores an Incredible 12,500 Points in Cinebench Multi-Tasking Benchmark

 

The chip was tested in Cinebench R15 multi-thread benchmark and the chip scores an astonishing 12,587 points which are beyond anything current-generation processors can achieve. AMD Ryzen Threadripper 2990WX scores around 5500 points in the same benchmark with 32 cores and 64 threads. The score we are looking at in the leaks shows more than twice the performance of the flagship Threadripper SKU. There’s also the EPYC 7601 SKU which scores around 6000 points that is due to the octa-channel memory support compared to the quad channel on the Threadripper CPUs.

AMD-EPYC-Server.thumb.jpg.c0dd62be281e9e09693acfc0ce84927a.jpg

  • Like 1

Share this post


Link to post
Share on other sites

Judging by early reports/rumours on engineering samples apparently the Zen 2 3700x is going to be hitting speeds of 4.5 ghz or higher.  I'm looking forward to that.  I held off on the 2700x because AMD has a pretty strong release schedule and I knew the Zen 2 stuff was coming relatively quick on the heels of 2nd gen.  Currently I'm running a 1700 so it should be an interesting upgrade if I go for it.

Share this post


Link to post
Share on other sites
On 10/5/2018 at 8:39 AM, Falconius said:

Judging by early reports/rumours on engineering samples apparently the Zen 2 3700x is going to be hitting speeds of 4.5 ghz or higher.  I'm looking forward to that.  I held off on the 2700x because AMD has a pretty strong release schedule and I knew the Zen 2 stuff was coming relatively quick on the heels of 2nd gen.  Currently I'm running a 1700 so it should be an interesting upgrade if I go for it.

Yeah. I upgraded from a 1700X to a 2700X. I couldn't get more than 3.9 on the 1700X, and often had to run it 3.8 to keep it as stable as possible. The 2700X has been quite a bit better, 4.1 is no problem. It will even overclock on it's own, to around 4Ghz, if I leave the clock speed at factory (auto) settings. I had been thinking about moving up to the 2950X which is basically  two 2700X's. Probably will wait until the 7nm chips are on the market, full on.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×