AMD's Ryzen CPUs (Ryzen/TR/Epyc) & Vega/Polaris/Navi GPUs

But, as is found in other segments, sometimes the lower cost product is actually better. I've bought plenty of overpriced POSes in my life. But, contra, I've also purchased cheap POS products when I should have just got the better one. It is an example of buyer beware. Not everything with a premium price is worth the ask, nor of the quality claimed. And there are times when the cheap will screw you worse than you think. Unfortunately, advertising only further obscures this, as does biased reviews. This is why we rely, often, on people we trust and know that have used it more than on what we see.

 

Certainly, and I'm not discounting this effect at all. It was mostly to point out that the performance may not always be related to the price because of differing priority.

 

You must be joking, right?

See:
https://www.cpubenchmark.net/compare.php?cmp[]=981&cmp[]=1011&cmp[]=3042


See:
http://techreport.com/review/32743/amd-ryzen-7-2700u-and-ryzen-5-2500u-apus-revealed/3


First off; I don't know any P9650 cpu's... Second, the AMD Ryzen 5 2500U is roughly on par with the current i5-825u0 and/or i7-8550u (see second link above).

With over 4x the performance (multicore) than the Core 2 Duo you were rocking an eternity ago and over 84% better single core / responsiveness too - by your thinking we should be paying more than the $800 asking price today. This is ignoring the huge reduction in TDP for todays platforms and the fact that $$$'s of 2008 are worth much more than $$$'s of 2017.

This is how things go; inflation ensures things seem (on the surface) to get more expensive. Even when they're not.

Yeah; I'm ignoring the gpu - because. As long as I can see the output on a screen; a dgpu, dGPU, DGPU or igpu is all equivalent to me (put the weighting you want on this component - but I'm sure today's offerings will blow away the 9600m GT in all things that matter for a mobile user).

The other thing to consider is that circa 2008 the economy and the industry was at a different place than they are today. The deals they were able to offer or were 'forced' to offer is much different than what they need to do today to survive.

That is one reason that your GL702ZC seems like such a good buy vs. this one to you. But that is not a valid comparison for many reasons... the biggest one being the poor and abysmal performance on battery and battery life - not to mention the noise of the system too.

That $805 system today that you're whining about with over 5x the RAM, 4x the CPU and (for $130 extra...) based on an SSD - even if the capacity is smaller (upgrade as you need...) - is so far past the 2008 Acer you hold in such high regard that it may as well be from a different planet.

And we can get it today for (much) less than a grand.

So, tell me; who are you trying to fool?

Manufacturers aren't getting away with anything that consumers don't let them do in the first place. Just because something is offered, it should not be obtained.

For the under a grand, fb/light gaming/sb's latte drinking crowd? This has success written all over it.

In any event; the past or present deals (i.e. sunk costs) you may have got has nothing to do with what you will be offered in the future (this future).

That is controlled by the market and hinted at by MSRP's.

I have no doubt that we will see this product again at the $600 range or lower soon enough - and at that point the comparison to 2008 tech or will be even sillier.

Don't dismiss the low end offerings because your requirements/tastes have gone upscale.



Take all the above (inflation, TVM and performance/battery life improvements) and come try to tell us how anything we can configure today isn't worth the cost.

You don't need to: there isn't anything you can present that would convince me or anyone of that point of view. The world keeps moving on.

The litmus test for the above statements is the fact that manufacturers are offering orders of magnitude improvements vs. decade old tech, overall. If they weren't? We'd all still be rocking core 2 duo's while Intel, AMD, NVidia, etc. would be 1/10th the size they are now (in $M/earned and employees, etc...).

I agree with you that some of it is artificial inflation.

My comeback; never buy anything at full price (MSRP). Which goes back to buying the best (highest performing) platform I can whenever I'm in a position to buy. Which allows me to wait for the best opportunity to get the 'next' latest tech at the best price at any given future time when I'm ready to buy once more.

The label 'low end' vs. 'mid range' doesn't matter anyway. That is just a constantly moving target - not only for the O/S, programs and games - but for the individual person's needs too.

Don't get so hung up on the absolute price. Look at what is actually offered in exchange. And I don't mean just compared to random past products. Nah...

What you always compare to is what are you able to do with your current system vs. what the new platform will actually allow you to do (better).

If you're already above the Ryzen 5 2500u - look elsewhere. If the R5u allows you to go above your current reach; whether it is a good deal depends on how much actual 'better' is offered vs. what you can afford to buy otherwise. If you're able to spend $130 more for an SSD (as an example) then the 'performance'/snappiness gained is worth it.

The cheapest (absolute $$) is not always the cheapest in the long term. Time is $$$$ (TVM) and (and but!) it is the one thing we can't earn - no matter how much we'd like to.

While we all would love to see these as $500 or lower offerings we can pick up at the local convenience store checkout line, the reality is much more complex than that.

Shareholders, staff and suppliers all need to be paid their fair or unfair share.

I wouldn't want it any other way.

Why? Because today isn't the day that will mark the last computer I'll buy. We all need them there, innovating and improving, for the time when a new system is needed once more.

Take care.

 

My point was that a mid-range laptop with an MXM dGPU in 2008 could be found for £600-£800
Today, OEM's are asking of consumers to pay the same for an ENTRY level system performance which is not even mid-range caliber and doesn't even have a dGPU - I don't care how much more powerful the new system is in relation to the old one.
What's next?
In another 10 years, instead of seeing excessively low cost of new technologies (as it actually should be today), we will see entry level systems being charged TWICE the price tag?
With mid-range quadrupling in price?

That defies even vaunted 'capitalist thinking' - but at the rate things are going, it seems to be where things are going in USA... and people actually think this is acceptable in a day and age of ever rising living expenses, stagnating wages, job losses due to massive automation, etc.?

I'm hardly discrediting this APU's capabilities... it's really powerful, but from a performance point of view it still falls into the category of 'entry level' and my point was that the price being asked for an APU system like this (regardless of its features) is beyond RIDICULOUS in 2017.

The economy changed, yes, but it doesn't benefit the consumer at all.

 

You may not care about the performance difference - others do.

You are also ignoring other, just as important, factors too:

Here's a simple example for you to understand:
See:
http://www.x-rates.com/average/?from=GBP&to=USD&amount=1&year=2008

See:
http://www.x-rates.com/average/?from=GBP&to=USD&amount=1&year=2017

See how the exchange rate is different back then? You paid a whopping amount of equivalent USD$$ for your decade old system vs. the $600 USD going price of the system today.

2008 midrange vs. 2017 low end doesn't matter either. Not when the so called low end is 4x and 5x more computing power today...

You can try to dismiss these important factors as much as you want - the world will move on without you.

For everyone else; today is always the best day to buy tech - if you need it now.

The prices are more than reasonable and if your budget allows you to spend more - do so! That is what I always will recommend.

O/S's, programs and games in 2008 don't compare to the same in 2017. This is a different world - a moving target. This is why yesteryear's 'midrange' can't compare to today's 'entry level'. The target is fluid and the definitions are ever changing.

You are fixating on a single aspect ($) when that is and has never been a measure of something's worthiness.

In another 10 years, we will still be paying the same ~$1K for a decent system - except it will come with an order or two of more compute capabilities than anything we can dream up now.

And if that future system is entry level vs. what can be bought at the high end is irrelevant - it will still be worth whatever consumers at that time are paying for it for all the reasons indicated in my posts here.

If you're just slamming the fact that this is an SoC/APU... you're still wrong.

The form isn't the important part - otherwise we'd all have apartment complexes to house our house sized computers with 1MHz speeds - what is important is the compute power and features each new advancement brings us.

To state "...the price being asked for an APU system like this (regardless of its features) is beyond RIDICULOUS in 2017." and believe it shows not how greedy OEM's are.

Rather, it shows how much you still have to learn with this aspect of computing and life in general.

So please, stop going on about midrange 2008 and entry level 2017.

The details and nuances here matter. A lot.

And nobody has been given any guarantees that any change in the economy will benefit consumers. Nobody.

 

6, bro. If I'd have to put my money, I would say those will be 6 cores CCXs. Why? Because mainstream 6 Cores CPU would need one, albeit good, chip. Bad chips will go for R5 and R3.

Hmmm, 64 doesn't divide on 6... perhaps I would loose my money.

 

Ryzen 1700 in a laptop.

 

That's awesome, the Ryzen 7 in the Asus GL702ZC even beat the 8700k / 7700k destops in multi-core

With the RX580 it's gaming GPU scores are passable, but the real benefit of this laptop is the 8c/16t Ryzen 1700 CPU in a laptop

 

Ryzen Mobile’s Features You Don't Know About
Published on Nov 15, 2017
Ryzen Mobile has features you probably don't know about. Stay tuned...

 

One thing for sure... Prices will go up. SSD's, memory is already up Price of Silicon Wafer Continues to Soar

"SUMCO, which is a Japanese silicon wafer manufacturer, recently announced forecast of its performance for fourth quarter and estimated that price of 300mm wafer will be at least 20% higher than that from fourth quarter of last year. SUMCO is a leading silicon wafer manufacturer that is accounted for two-thirds of global silicon wafer markets along with Shin-Etsu Chemical."

“We are also planning to increase price of silicon wafer by 20% in 2018.” said CEO Hashimoto Mayuki of SUMCO. “ Price of silicon wafer will continue to rise in 2019.”

 

Bayonetta [HP Envy X360][720p][Ryzen 5 2500u][17.7]

Resident Evil 5 (w/ power)[1080p][40fps][Ryzen 5 2500U][17.7][OC stock]

 

I have a feeling that one is going to be VERY popular when it launches.

 

I already have it.
The RX 580 (despite benchmarks) behaves pretty much similar to a GTX 1060 performance wise and is capped to 68W.
Asus could have made better cooling for the GPU and avoid loud noise that accompanies it, but still.

I was wondering though what is Ryzen's 1700 lid (Integrated heat spreader - IHS) made of?
I am probably going to get another 16GB stick for this laptop this month (black friday) so I can get dual-channel on it, and while I'm in it, I will probably do a re-paste with GeLid Extreme.
I was also thinking on getting Conductonaut Liquid Metal obviously as its the best, but I need to know if this would damage ryzen's lid (the heat pipes/heat sink as far as I know are copper based though).

 

IHS is nickel-plated copper. Liquid metal will be fine on it.

 

Very useful. Thank you. I found similar info online, but it wasn't clear if the lid itself was made out of it or not.

 

The IHS is soldered with AMD CPU's, Ryzen 7 and TR, so no easy delidding.

Here's a delidding of the ThreadRipper IHS for your entertainment:

I don't recommend conductive TIM's, too much chance of droppings and drippings causing trouble. There are plenty of good non-conductive pastes that will do well enough.

I wouldn't re-paste at all though, why start out on that whole chain of re-pasting for a couple of degree's of temperature drop, especially when you aren't thermal throttling?

 

New AMD Radeon Software Crimson ReLive 17.11.2 driver (11/17/2017 ). Nothing mentioned.

And please remember laping your lid. AMD's newer lid is trash. It's heavy concave. Intel's lid go same way as well.

 

Sounds like Oculus broke it since they don't support it any more, so maybe AMD hasn't addressed it because the retail ones are still fine?

 

Will people get the truth? Nope. This can or will be easily moved under the carpet. From both.

 

Yeah, just speculating. I'm guessing after some findger pointing, AMD will eventually release a driver that works with it.

 

Firefox Hardware Survey Puts AMD Ahead Of NVIDIA In Terms Of GPU Market Share
https://wccftech.com/firefox-hardware-survey-puts-amd-ahead-of-nvidia-in-terms-of-gpu-market-share/

The GPU market share as reported by Firefox are as follows:

• Intel’s integrated GPUs leading at a solid 65.39%.
• AMD GPUs following at 15.40%
• NVIDIA GPUs close behind at 14.167%

Probably Intel Optimus comes into play, many of those Intel laptops also have AMD / Nvidia dGPU's, and Firefox is running on the iGPU due to the low loading.

 

This AIO is made for Threadripper, but can it handle it when overclocked?


Tear-Down: Enermax Liqtech Threadripper Cooler


I thought this tear down (Jayztwocents) of a water block running for a few months was enlightening, it had developed fractures that could lead to leaks and total failure down the road.

The fluid apparently cleaned up the internal metals nicely, but a cracked water block @ ~4:10 negates that for me

How well did the Primochill Vue fluid hold up?? Let's tear blocks apart and see!

 

AMD RX Vega 64 Gets 22% Performance Uplift In Wolfenstein 2 After Latest Patch
https://wccftech.com/amd-rx-vega-gets-22-performance-uplift-wolfenstein-2-latest-patch/

"Wolfenstein 2 is the first product since the announcement of a long-term strategic partnership between Bethesda — publishers of Fallout 4, Dishonored 2, DOOM, Wolfenstein 2 among others — and AMD earlier this year, promising to deliver “unprecedented performance for our fans” and it shows."

 

CES 2018 BEST OF INNOVATIONS: COMPUTER HARDWARE AND COMPONENTS
http://www.ces.tech/Events-Experiences/Innovation-Awards-Program/Honorees.aspx
AMD Ryzen Threadripper 1950X
Advanced Micro Devices, Inc. (AMD)
TW, Ven Lvl 2, Titian 2303/2304/2305
As a do-it-all powerhouse built to serve the needs of content creators and enthusiasts, AMD’s Ryzen Threadripper 1950X supports 16 cores and 32 threads, delivering world-class performance built around AMD’s powerful “Zen” architecture. "

AMD Ryzen™ Threadripper™ 1950X Wins CES 2018 Best of Innovation Award
High-Performance processor deemed most innovative technology in Computer Hardware and Components category
https://www.amd.com/en-us/press-releases/Pages/amd-ryzen-threadripper-2017nov09.aspx

AMD Is Ryzen Up!

 

Considering the last couple of generations, I would be tempted to say that most machines running an iGPU in this hardware survey that actually has a dGPU installed in them are going to be NVidia.

 

The 15z is on sale at HP for $574. http://store.hp.com/us/en/pdp/hp-envy-x360-convertible-laptop-15z-touch-1za07av-1

Seriously considering one. The estimated ship date is 12/19 though. I don't want to wait that long. I might see if Best Buy will price match.

 

That's another great deal, one I wouldn't worry about waiting on at that price. BB's models are usually unique enough they won't be able to price match. And, BB also has a "must be in stock" at competitor rule, so that negates pricing matching in this situation too.

 

Someone on Reddit said BB changed their model number so they wouldn't have to price match HP, but he finally got them to. They also told him they would have it on sale on Black Friday.

 

Well, I think the parts build out is usually different too, but maybe it is just a part number subterfuge, sometimes.

Yup, the HP sale is a "Pre-Black Friday" "Black Friday" sale, keep watching BB for their BF sale - only problem is they are usually limited to stock on hand, I'd get one now

 

I'm hesitant to believe anything like that coming from Reddit, but no harm in trying. Typically BB gets builds with different components entirely from other places to stop that sort of thing.

 

This is true that sometimes Best Buy carries notebook SKUs that are only sold by them. For example, my very first "gaming" notebook, the Asus G73JH, was a Best Buy exclusive config called the G73JH-BST7, which was cheaper than non-BB SKUs because it came with lower end specs like a 900p instead of 1080p screen, less RAM, less HDD space, barebones WiFi card, DVD intead of BD drive, etc.

 

Inside the Envy 5Z http://apusilicon.com/hp-envy-a-closer-look-at-the-internals/

The ram is easily accessible which is what I needed to see. I can get it a lot cheaper than the $150 HP wants for an upgrade.

 

AMD Ryzen and Threadripper Prices Over The Weekend Dropped Significantly
https://www.guru3d.com/news-story/amd-ryzen-and-threadripper-eu-prices-drop-significantly.html

"Update: we have received reports from the Nordics and the UK that the price drops are in effect there as well. USA Newegg is also showing lower prices. And a report from Australia just came in as well, this is a worldwide price drop. We asked, but AMD has no comments currently on what we are seeing. "

http://www.microcenter.com/search/s...683&NTX=&NTT=&NTK=all&page=1&sortby=pricehigh

https://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007671 601295133&IsNodeId=1&bop=And&Order=PRICED&PageSize=36

Update: Checkout @Mr. Fox 's awesome list of Ryzen / ThreadRipper sales links
http://forum.notebookreview.com/thr...ners-welcome-too.810490/page-32#post-10635363

 

That's nice to see. The practivce of putting RAM on the opposite side of the mobo from the access panel cannot end soon enough.

 

Totally agree. If the ram is not soldered in, then it needs to be user accessible. When I had a stick of ram go bad some months ago, I sent my laptop to MSI and had them remove it from under the keyboard and put it in the slots where I could get to it.

 

If they just left keyboards removable, then you could access both sides. Clevo has kept that design feature, always disappointed when I see systems without it.

 

Significant Sales on AMD Ryzen & Threadripper CPUs, Upwards of 30% | Black Friday
By Steve Burke Published November 21, 2017 at 5:28 am
https://www.gamersnexus.net/sales/3132-huge-sales-on-amd-r7-1700x-1600x-and-threadripper-1950x

Spoiler: article text...

"In the calm before the global celebration of consumerism, it would seem that the entire range of AMD processors has gone on sale – or most of it, anyway. Several of these have tempted us for internal machines, at this point. The Threadripper 1950X has been available as low as $800 (from the usual $1000) price-point, the R7 CPUs are cut into R5 prices -- $260 for the 1700X is now common, and R5 CPUs have also been dropping in price. The timing is excellent, too, as we just posted our Best CPUs of 2017 Awards, which include several of these sale items.

Threadripper 1950X (launch MSRP $1000): Let’s start with the big one: The Threadripper 1950X CPU, which we reviewed originally here and for which we’ve updated game tests here, usually retails at $1000 MSRP, but is now $800. We had already declared Threadripper to be an exceptionally fierce competitor to nearby Intel i9 CPUs – primarily the i9-7900X – but at $800, it’s a no-brainer for users of heavily multithreaded workstations. Anyone working with rendering workloads that lean on the CPU would be wise to consider the Threadripper 1950X. One use case would be data sets with large vectors or tile-based renders, particularly those with short per-tile render times (benefiting the CPU tile-count over fewer GPU tiles).

Recommended motherboard for TR 1950X CPUs: If you’re looking for a motherboard to go with it, we can recommend the Gigabyte X399 Designare or ASUS Zenith Extreme.

AMD Ryzen R7 1700X (launch MSRP $400): Although the 20% price reduction on Threadripper was impressive, the R7 1700X has undergone precipitous drops to $280 (and lower) lately. Presently, at the time of this writing, the R7 1700X is priced at $280 with the cooler, and is normally priced at its $400 MSRP. The R7 1700 is $10 cheaper at $270 and, if you’re overclocking, is basically the same thing. Save the ten bucks. If not overclocking, the 1700X will give higher out-of-box clocks, and would be worth the $10. At $400, we never really liked the 1700X – we though the 1700 made far more economic sense, and that a 10-minute overclock would counter the difference. At $10 more than the 1700, as opposed to the usual $100 increase, it’s much more sensible to recommend – almost crazy, really, and is a testament to the MCM configuration that allows AMD to shuffle the same Zeppelin dies between all CPUs. If you were already in the market for an R7-class CPU, now seems to be a good time to jump on one. If you’re not sure if Ryzen is right for you, check our CPU Awards guide.

Recommended OC motherboard for the R7 CPUs: Our AM4 motherboard guide goes live at 3PM EST today, but if you’re looking for a board to go with the R7 1700X, we can recommend the Gigabyte X370 Gaming K7 as a high-end overclocking option. The K7 has a clock gen with 1/100th MHz accuracy, a strong VRM, dual-BIOS, and a good price. That’d be well-suited for the R7-class CPUs.

AMD R5 1600X (launch MSRP $250): The R5 1600X, which received an Editor’s Choice Award and whose non-X counterpart was given the Best Overall award, is presently available for $200 via Newegg and Amazon alike. The R5 1600(X), we think, offers tremendous value over competing i5 CPUs. Where the i7-8700K and 7700K remain leaders in raw gaming performance, particularly the 8700K, the R5 1600 and 1600X stay much closer to i5 CPUs in gaming workloads. Coupling this with the fact that (1) 8th-Gen i5 CPUs are presently overpriced, and (2) non-Z370 motherboards aren’t yet available, and you’ve got the R5 CPUs leading the mid-range class. We like the R5 1600(X) for its versatility as an entry-level production CPU and well-suited gaming option.

Recommended OC motherboard for the R5 CPUs: To OC these CPUs reasonably, you could get by well with either the Gigabyte X370 Gaming K5 ($135, includes clock gen, but slightly weaker VRM) or ASUS X370-Pro.

Check back later today for our Best AM4 Motherboards guide.

- Steve Burke

Here's Why You Should Choose AMD this Holiday Season

 

Asus Outs ‘New Upcoming Processors’ With AMD B350/X370 BIOS Updates
by Derek Forrest November 20, 2017 at 4:45 PM
http://www.tomshardware.com/news/new-amd-processors-asus-bios,35969.html

"Is AMD readying a new round of processors for its AM4 platform? It certainly seems so, with Asus releasing a new BIOS update for its entire range of B350 and X370 motherboards that effectively confirms this.

Asus released a series of BIOS updates for its full selection of AMD B350 and X370 chipset motherboards. Each of the individual updates share the same description:

“Update to AGESA 1071 for new upcoming processors.”

We’ve already discussed the finer points of the AGESA protocol, and it’s not a new or unexpected feature for the AM4 platform; AMD and its partners frequently release BIOS updates that support new revisions of the code. However, Asus stating that the update is for new upcoming processors indicates that AMD will be expanding its AM4 processor family in some way. Asus currently seems to be the only AMD motherboard partner offering the new BIOS (dated 11-17-17), and it seems the company has let a significant cat out of the bag."

GLOBALFOUNDRIES Introduces New 12nm FinFET Technology for High-Performance Applications
https://www.globalfoundries.com/new...-technology-for-high-performance-applications

AMD Transitioning To 12nm LP Process For Vega, Ryzen In 2018
http://www.tomshardware.com/news/amd-ryzen-vega-12nm-lp-2018,35502.html

AMD’s CTO on 7nm, Chip Stacks
https://www.eetimes.com/document.asp?doc_id=1332049&page_number=1

ASUS Confirms New Zen CPUs | Analysis of 12nm Performance of Pinnacle Ridge
Starts transition at about 04:15

 

FINAL ANSWER! - GTX 1060 vs RX 480 & 580

$0 RX 480 UPGRADE! - How to Get More FPS For Free!

 

Too much risk and better value is OC'ing the hell of Radeon Cards. I was running a 5750 Radeon with 5770 clock rates which was very stable.

 

Our first Ryzen 5 2500U benchmarks are in and Intel has every reason to worry
As fast as a Core i5-8250U but with more than double the graphics performance, the Ryzen 5 2500U is launching with just the momentum AMD needs in the notebook space.
by Allen Ngo, 2017/11/25
https://www.notebookcheck.net/Our-f...Intel-has-every-reason-to-worry.266618.0.html

 

That's a good initial showing but 22% slower (uh... CB... whatever) than the i7-8550U (single core) isn't something I would give up. That's like going back to 2013...

Anyone that buys a ~$600 platform to run CB or similar high core count workloads (I don't) may find this ideal, but for anyone else? Intel isn't even beginning to sweat yet, imo.

I'll be sure to check out the full reviews when they're available and also looking forward to the Ryzen 7 mobile reviews too.

 

LOL, the review is stating against a 8250, not 8550. With intel is the answer always spend more for something faster? if so that is not the market the CPU's are targeted for.

 

The AMD Ryzen mobile 2700u compete's against the 8550, there are benchmarks from AMD that show it's also up to the task at besting Intel's new 8th gen ULV releases, and beyond.




Now we wait for the AMD Ryzen mobile 2700u to start showing up in products

 

XFX launches Radeon RX Vega 64 and 56 Double Edition
Published: 24th Nov 2017, 10:44 GMT
https://videocardz.com/74175/xfx-launches-radeon-rx-vega-64-and-56-double-edition

" XFX makes custom Radeon RX Vega official
XFX today launches two cards, both featuring the exact same design. According to official specifications, XFX’s Double Edition have dual 8-pin power connectors, which means earlier pictures showing 8+6-pins were likely showing an early prototype.

XFX has only released four pictures of the new cards, none are showing the PCB, so we don’t know how far the customization of power delivery has gone. The board is custom, shorter than original design with power connectors placed in a center (I wonder how will this affect cable management).

XFX RX Vega Double Edition features four display connectors: three DisplayPorts and one HDMI 2.0. No word on the pricing or availability yet."

XFX Radeon RX Vega 64 8GB Double Edition (RX-VEGATDFF6) Graphics Card
https://videocardz.net/xfx-radeon-rx-vega-64-8gb-double-edition

XFX Radeon RX Vega 56 8GB Double Edition (RX-VEGALDFF6) Graphics Card
https://videocardz.net/xfx-radeon-rx-vega-56-8gb-double-edition


PowerColor Radeon RX Vega 64 Red Devil: real custom Vega
The first unboxing of PowerColor's new Radeon RX Vega 64 Red Devil is here
https://www.tweaktown.com/news/59922/powercolor-radeon-rx-vega-64-red-devil-real-custom/index.html

Powercolor Vega 64 Devil


Powercolor RX Vega 64 Devil Preview
https://overclock3d.net/reviews/gpu_displays/powercolor_rx_vega_64_devil_preview/1

 

The review(er) can compare to whatever they want. Doesn't make it a good or even valid comparison. Just because something performs similarly doesn't mean it's equivalent.

I compare to what is available to me, today. Intel's answer isn't to always spend more; that is life.

You pay and you play.

With your reasoning it is better to pay 5% less (total cost of completely setup platform) and get 22% less performance. Okay...

Intel, AMD and any other manufacturer can market the cpu's for whomever they feel is best served by their products. I, on the other hand, am not limited to such ploys.

I simply use the best products available for my specific needs; regardless of whether I'm the target market or not.

I'm fairly confident I'm not the only one with this ideal either.

 

your logic would dictate that just one system be offered to everyone, this is not how the market works. come back with a valid argument if you can find one.

Intel/AMD are not worried about your single decision to buy an 8550. I will be the last to try and change your mind either. The end choice ca be personal but I do not again discount the choice either. I can easily see where either is a valid choice for many, that you can't is to your detriment.

 

AMD: Ryzen 3 1200 delivered with 8 active cores-notebookcheck.com

"Just a few weeks ago there were reports that apparently Ryzen 5 processors with 8 active cores + SMT were delivered, although actually only 6 cores should be active. Now there are apparently similar incidents on some Ryzen 3 processors sold in Russia."

 

Patch NPT ON Ryzen for better performance
https://level1techs.com/article/patch-npt-ryzen-better-performance

Spoiler: What is This?

What is This?
The golden age of Ryzen for gaming virtualization is upon us. If the reader is not familiar with this technology, it allows one to run a virtual machine with direct access to a secondary graphics card. This allows full GPU acceleration of workloads running inside a virtual machine. The most common use cases for non-server non-enterprise uses are for running legacy applications and games that require direct access to a GPU. If everything is working correctly, there is a negligible performance hit. Running a gaming VM vs “bare metal.”

On Linux, with the launch of Ryzen from AMD this type of workload had been problematic. At first the platform was not well suited for sharing PCIe resources between host and guest. This turned out to be a software oversight and was fixed within a couple of months of the launch of Ryzen via an AGESA platform update from AMD. However, the performance has not been as good as it has been on Intel platforms and this was down to a software bug with KVM, one of several hypervisors available for Linux that enables this type of virtualization.

As of this article, that bug has been fixed (mad props to Geoffrey and Paolo) but it will be some time before the patch is included in mainline Linux kernels.

This article will walk you through patching your kernel if you are using Ubuntu, Debian or Fedora distributions. If you are on Arch, you probably already know what you are doing and all you need is the patch itself, which is here.

You should also know that this essentially completely fixes the issue on Ryzen CPUs, but Threadripper (being a bit newer) still has a couple annoying issues that I am certain will be worked out with software fixes. In fact, one reason I am using Ubuntu as the basis of this article is because I am working with a board partner to nail down the specifics of these issues (at least on the hardware side).

Spoiler: Getting Started

GETTING STARTED
You have a choice about what kernel you use. This patch will work on many different kernels including shipping kernels with Debian, Ubuntu and Fedora. If you wish, you need not replace your entire running kernel – the patch is actually just for kvm_amd.o (which is, 99.9% of the time, just a kernel module on your distro).

If you wish to run a bleeding edge kernel, which is 4.14-rc6 as of the time of this article, then you can clone the kernel source from a source that is appropriate for you.

UBUNTU (BLEEDING EDGE)
git clone git://git.launchpad.net/~ubuntu-kernel-test/ubuntu/+source/linux/+git/mainline-crack v4.14-rc6

UBUNTU (JUST MODIFY WHAT YOU HAVE)
apt source linux-image-$(uname -r)

NEXT STEPS
Navigate to the kernel source folder on the CLI

Copy your current kernel’s options and config from /boot to this new unconfigured kernel

Apply the patch

The commands might be something like

cd v4.14rc6
cp /boot/config-4.10.0-37-generic .config
patch -p1 < ryzen.patch

Where ryzen.patch could be /home/youruser/Downloads/ryzen.patch (or you copied ryzen.patch there), or whatever.

https://patchwork.kernel.org/patch/10027525/

Just copy-paste this part into a ryzen.patch file (copied from frame on site, not sure if it's CR/LF's are preserved correctly, so get it from web site if available.):

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index af256b786a70..af09baa3d736 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -3626,6 +3626,13 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)
u32 ecx = msr->index;
u64 data = msr->data;
switch (ecx) {
+ case MSR_IA32_CR_PAT:
+ if (!kvm_mtrr_valid(vcpu, MSR_IA32_CR_PAT, data))
+ return 1;
+ vcpu->arch.pat = data;
+ svm->vmcb->save.g_pat = data;
+ mark_dirty(svm->vmcb, VMCB_NPT);
+ break;
case MSR_IA32_TSC:
kvm_write_tsc(vcpu, msr);
break;

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index af256b786a70..af09baa3d736 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -3626,6 +3626,13 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)
u32 ecx = msr->index;
u64 data = msr->data;
switch (ecx) {
+ case MSR_IA32_CR_PAT:
+ if (!kvm_mtrr_valid(vcpu, MSR_IA32_CR_PAT, data))
+ return 1;
+ vcpu->arch.pat = data;
+ svm->vmcb->save.g_pat = data;
+ mark_dirty(svm->vmcb, VMCB_NPT);
+ break;
case MSR_IA32_TSC:
kvm_write_tsc(vcpu, msr);
break;

The problem this fixes is quite funny -- basically, in the AMD version of this patch, the SVM properties for cachable operations are ignored. Cache was never used, up till now. So of course performance with KVM was not great. Some people, myself included, opted to use the Xen hypervisor in lieu of KVM which did not have this bug (and performance there was pretty good). Xen has its own sets of headaches, however, and I prefer KVM.

You should see “Hunk 1 of 1 succeeded”) or something similar to that. (It is OK if you get a message about an offset.)

Now, if you are intending to replace your running kernel with a new version, you must build the kernel, install it and reboot.

The commands I used on this Ubuntu install were:

yes '' | make oldconfig
make clean
make -j 16 deb-pkg LOCALVERSION=-custom

Note the –j 16 is the number of threads you want to run for the compile job. I was on Threadripper so 31 or 32 might have been appropriate.

Did you get a segfault on compiling? You may need to update your UEFI or RMA your CPU. More info here. This process will typically take between 4 and 45 minutes depending on your system.

Once the compile is done, you will have a number of .deb binary packages. These must be installed with a command such as

sudo dpkg -i linux-headers-4.14.0-rc6-custom_4.14.0-rc6-custom-1_amd64.deb linux-image-4.14.0-rc6-custom_4.14.0-rc6-custom-1_amd64.deb linux-libc-dev_4.14.0-rc6-custom-1_amd64.deb

If your system was already configured for passthrough virtualization, then you can reboot and test out the new kernel. If not, please see one of our earlier guides on setting up passthrough virtualization.

UPDATING KVM_AMD.O ONLY
If you’d rather not replace your entire kernel, the steps are slightly different. Instead of installing the new packages with dpkg, you will simply copy the kvm_amd.o kernel module over top of your old kvm_amd.o (be sure to save a backup if you want to undo). A command such as:

cp arch/x86/kvm/kvm-amd.ko /lib/modules/$(uname -r)/kernel/arch/x86/kvm/

will overwrite your kernel module for kvm-amd in / with the one you just compiled in your kernel source working directory. From there a simple

rmmod kvm-amd
modprobe kvm-amd

should remove and reload the kvm kernel module (note that no VMs can be running. If you are unable to rm the module, then simply reboot your machine for the updated kernel module to be installed). Note that my assumption is that npt=1 is the default on your system. If you have added module options in /etc/modprobe or equivalnet to disable npt (npt=0) then you will need to change that to npt=1 to re-enable nested page tables on your system.

CONFIGURATION RECOMMENDATIONS

I recommend

iommu=pt amd_iommu=on

for your kernel parameters in /etc/default/grub for both Ryzen and Threadripper platforms. It may also be necessary to enable unsafe interrupts for the vfio_iommu_type1 kernel module. To learn more about the particulars and step-by-step to setup this type of virtualization, please see our earlier articles for the step by step. Allowing unsafe interrupts by passing the option to the kernel module is likely the only thing you might have to do beyond those instructions.

WHERE DO WE GO FROM HERE?
Our tested hardware config is a 16-core Threadripper 1950X, Gigabyte Designare X399, GSkill Trident Z DDR4-3200 B-Die memory, ASUS Strix Fury (Host), and a Vega 56 (Guest). We are using the Noctua 120mm TR4 Tower Cooler for cooling and it does a marvelous job.

If you are bored, post your system specs and your before/after benchmarks or framerates when applying the patch. One can never have enough data!

We will be doing a full AMD guide/article once Kernel 4.15 drops. (Update: the patch didn't make it in time to release in 4.15)

It has a lot of patches/improvements for everything in the AMD ecosystem. I would encourage you that, if you have bought a Threadripper system, and you are having trouble getting your passed-through peripherals to work, that you kindly let AMD know about your hardware configuration so it can be tested:

http://www.amdsurveys.com/se.ashx?s=5A1E27D24DB2311F

I have been working very hard behind the scenes on the issue with board parners, and progress is being made. Right now Vega 56/64 works well as a guest GPU (but you do have to reboot your host machine -- I believe I've seen a fix for this slated for inclusion in kernel 4.15, but I have not yet tested it.)

Livestream: Fix for Ryzen NPT Applied & Tested

---------------------------------------------------------------------
Livestream: Headless PCIe Passthrough? Coming Very Soon


A little teaser of what is to come
https://forum.level1techs.com/t/a-little-teaser-of-what-is-to-come/121641

 

Sorry, but your conclusion (below) proves that you can't have a logical line of thought with me.

I too can see where the choice of either system is valid for many. But that is not the conversation we were having.

I commented on the fact that Intel had 'reason to worry' (not).

I then commented on your assumption that who the products are marketed to is who buys them. Also a false assumption.

Now you're trying to twist this into me wanting one system for everyone? Sorry, but I see through your circular reasoning here.

If you really can't see how you're trying to twist my words to make me look dumb is to your detriment, not mine.


Btw, this is on topic because that is what I was originally discussing (on topic). Just because you happen to not agree is okay with me, but you can't just decide this is ot and delete other's comments as you feel like while still leaving your comments in the thread that cast shadows on other's ideas and thoughts.