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

They said the official announcement is coming in the next couple days. If you own a ZE Monoblock, they have confirmed a free upgrade kit will be shipped. Have not clarified on the other blocks yet whether it will be free or cheap. Here is the facebook link:
https://www.facebook.com/EKWaterBlo...1836.182927101761329/1593016050752420/?type=3

 

AMD Raising GPU Production!

My 10 Year Old Chiller WORKS! -20C on THREADRIPPER!

 

I at least would have liked to know if you went through all that trouble what kind of extra overclock could be as the result?

 

Hopefully Linus will come out with a video on using the chiller for OC tuning.

Although it's possible Linus could get some more OC headroom at -20C, others have taken LN2 level cooling on Threadripper and gotten better OC's...here's some LN2 tuning hints:

Liquid Nitrogen Overclocking with Threadripper @ 5.2GHz

 

My main interest here is that Zen+ at 12nm may give the same or better results in the end.

 

TR4 has 4 years of new CPU supported life according AMD, allowing us to replace the current TR CPU with one of the new Zen+ CPU release(s), and you could use the same chiller cooling solution in the same motherboard (or a new TR4 motherboard if you like) moving forward. There might be more or less headroom at -20C with the newer Zen+ CPU's.

Then we can do this again with TR4 Ryzen 2.0, and maybe even again with TR4 Ryzen 2.0+, for a total of 4 CPU release's using the same TR4 socket.

That's a nice long run to amortize an extreme chiller cooling solution.

 

That's one reason to love AMD among many. Multiple generations on the same socket has always been one of their strong suites. Intel has had how many core processors, and how many socket changes? Can you say chipset/motherboard gouging? I am sure they could have used the same socket for at least the last several generations. @ajc998 could probably answer that question

Sent from my SM-G935T using Tapatalk

 

The reason is they have less money so they need to design a mobo that can last long w/o re-design. In a way its good.
I think Papusan, Mr Fox, Phoenix and countless have upgraded to skylake/kabylake once they are out.

 

Guess this is all still rumored but only 200 MHz is not enough to generate any excitement on Zen+ 12nm. An additional 500 MHz or at least 10% might have been but then again it is still early too.

https://www.bing.com/videos/search?...02490C18C093E8D38E0102490C18C093E8D&FORM=VIRE

 

If 200MHz is all you get, I would call it a big fiasco.

 

Even a 400-500MHz increase will not give them the performance crown but would put the CPU's at a much better showing against Intel offerings. The other problem will be everyone who believed in the hype before of Ryzen improvements will loose faith in the hype of 7nm. This would not bode well for AMD stock after the lackluster Vega release.

 

Whatever improvements there are in Ryzen+ we will know soon enough when the products are released so reviewers and owners can publish what they see from the actual production products.

There is no reason to put expectations out there when no official AMD news has been published. Your habit of doing this before every AMD release isn't insightful, it's inciting...

Why do you feel you need to poke holes in imaginary balloons?

Rumors of AMD performance improvements by 3rd parties are often over the top, and everyone should ignore those and wait for the actual numbers discovered upon release.

Also, you can't draw conclusions between two products that are based on different architecture + different die process. Those are 2 of the most divergent parameters out there, the exact opposite in possible ties or connections for predictions.

Improvements between Ryzen 1.0 and Ryzen 1.0 + won't carry over to Ryzen 1.0+ to Ryzen 2.0, they aren't comparable. We can't make any conclusive estimations about Ryzen 2.0 based on Ryzen 1.0 vs Ryzen 1.0+ improvements.

 

I think it's ok to speculate on future performance of upcoming chips based on leaked information, or even just pure speculation based on theory. No one's saying that these are 'facts', it's just speculation, however valid or invalid that may be.

 

Not really, especially with the AMD antagonists putting out ridiculus speculation on how far ahead AMD will be - to put out impossible expectations - so that when AMD delivers, no matter how cool it is, it doesn't meet expectations and then the antagonists say AMD failed.

It's aggravating to see AMD sucker punched with these trolls preceding every generation, with the same negative result.

There are no useful leaks, it's all BS.

 

The issue is for everyone out there looking to inflate there are those of us out there living in the real world. You blame everyone else where AMD itself could easily come out and put the scores straight so to say. Intel came out way ahead of time for up to the 7980xe and performance expectations.

AMD's own graphic shows 12nm closing the gap to Intel's 14nm and it looked like it was supposed to close it substantially. 200MHz is no where near substantially!. So you can blame AMD for this so called talk.

 

That's the other BS quote you make, calling out AMD to dispel the garbage "leaks"... and how AMD are "blowing it" by letting those garbage "leaks" go unanswered.

You need to think about that more... really it's silly.

Would you really expect AMD to drop what they are doing and respond to some random nerd posting BS about their unreleased products?

Of course not, that's ridiculous. If you don't understand this, think about it some more, because it's totally obvious why AMD wouldn't do this.

AMD isn't going to release hard information on product under development, and neither is any experienced developer or company, it's just a bad idea.

There is nothing to be gained by it.

AMD can't release any product information to refute baseless claims made in garbage "leaks" because AMD aren't releasing information on unreleased product.

It's that simple.

 

If the supposed leak is based on false information in a database, yes it would be to AMD's advantage to set the record straight. If it were just a random false statement with no data backup, then yes no response is needed.

 

It can often be a very good idea to drop out small "leaks". The fall will not be that big when the launch comes. High expectations who crash can cause greater damage.

 

AMD may have only a 200mhz bump, as they are well known for this tactic, but if thats the case, I almost guarantee the overclocking headroom will be substantially more than gen 1. The same crap was said just before Phenom 2 was released, about not needing to upgrade to it because the stock clocks were not much better. We all know how that went after realease. Phenom 2 was a beast of a clocker. Like I keep saying, I think this is the plan, as I will stay optimistic about zen+. I would definitely purchase a 2700X or a 2800X if they clock to 4.5 or above. I expect Zen 2 to hit 5ghz overclocked

My 1700X does not like 4.2 for any reason whatsoever, no matter what I throw at it, it refuses to go that high for any significant amount of time.

 

AMD would be idiotic to respond to leaks. Buffoon's of the first order. The only way to dispel a leak's "false information" is to publish "true information."

AMD is not going to release any information on unreleased products, why don't you understand this most basic business precept?

I think we've all seen that the "best" leaks, with the most awesome results and therefore expectations, those are the ones that are total BS.

The "small" leaks, the "true" leaks of early production or engineering samples are rarely as exciting, and since they are usually "disappointing" due to early software and hardware tuning, do as much or more damage as the "big whopper lie-leaks".

And, then we have the two (truth and falsehood) mixed together in a clever ruse to confuse...

Ryzen’s Desktop APU Beats Discrete GPUs!

Disappointing if true, Intel has faster hybrid's, damaging either way without confirmation - muddling along wondering until it really releases for confirmation. Why buy the AMD hybrid if the Intel Hybrid is "faster"? Damaging all the way around. Lousy BS leaks, always out to damage AMD's success.

That's how the "opposition" does it's damage to AMD, poking holes in the surprise at release, taking the wind out of the sails of success or adding anchor weights to the reality of "a minor upgrade", assuring some kind of disappointment no matter how it ends up.

Anti-AMD troll's modis operandi are easy to spot, and therefore easy to ignore.

 

There's also the fact that the technical documents describe 12nm LP as designed for high performance/high clocks (whereas the 14nm LPP is designed for low clocks and mobile parts in general).
200 MhZ increase over base doesn't correspond with 10% performance increase... and I fairly doubt that IPC increases would have happened for a simple refresh (at least, we hadn't until now) ...
This process should theoretically allow AMD much higher increase than 10% in clocks alone and maintain existing TDP.
If AMD was going to 12nm designed also for low power, then 200MhZ (or even 300 Mhz) bump up might have made sense... but considering we're talking about a high performing process, then something doesn't mesh well.

So, the 200 MhZ increase we are seeing in those early reviews could still be engineering samples (which hasn't been verified of course). There's a chance we might see much higher clock rates... unless they are mitigated by the premise that AMD decided to integrate an iGP with Ryzen refreshes (though this seems more likely for Zen2 on 7nm)... or if this process definition of 'high performance' simply means something else... or is more suitable for GPU's and doesn't translate well for CPU's?

 

The problem is, but not always, that it is true the end results from AMD have some level of disappointment. Now if AMD would publish some true information that would dispel the hype then with a lowered, or properly adjusted, expectation the products would be better received.

Intels hybrid solution looks to have a higher TDP so between that and the higher IPC there is a good chance it will be a better end solution.

 

The FASTEST* Laptop We’ve Ever SEEN

 

I wouldn't be so sure.
To date, AMD has been on the mark with performance numbers from what they were expecting of Ryzen... and actually managed to go beyond those expectations.
The 10% performance increase comes from various sources and technical information on 12nm LP... which is touted to be an increase in comparison to 16nm FinFet process that Nvidia used for Pascal (which is already a process designed for high performance and high clocks).
Officially, AMD hasn't finalized clock speeds nor advertised how much of an increase we can get or expect from the refresh... so everything is still up in the air.

Also, the 'higher IPC' on Intel part is only 5%. Not really much of a difference.
The main issue is that Intel got basically an entry level GPU level of performance with that Vega they got from AMD... but, bear in mind that AMD will also release it's own versions with this Vega chip most likely.
Plus, on the AMD's end, the 2500u and 2700u laptops already are comparable or better than Intel in the CPU arena while also quite far ahead in the iGP.
AMD could easily pair 2700u with say RX Vega 56 that's been undervolted and slightly underclocked (by say 10-15%).
Or, they can easily just make a cut down Vega 23 for example with clocks in the range of say 1100-1200 MhZ (which is efficient for 14nm LPP), undervolt it, and pair it with 2500u/2700u in a hybrid form like they did with Intel... or just keep that more powerful vega as a dedicated GPU.

 

My point is the results *aren't* disappointing, when you discount all the hype from "leaks" before release that promise ridiculous levels of performance, that we all dream about, but know aren't possible.

Without feeding the hype train before release, the results are just that, results.

And, when you have real production numbers to compare against other real production numbers, taking price and features into account, there is no need to be disappointed.

Without the hype, if you are disappointed, you set your own expectations too high, and only you can solve that dilemma.

 

Yeah, I see. Everything will drip down to "Bang for bucks". Then will all be happy Ryzen is a disaster for overclocking. And If AMD is able to push stock clocks higher, lets say 4 bin. We will probably still see the same

 

Correction, Ryzen is a disaster for overclocking on 14nm LPP past 3.9 or 4 GhZ, because the process is designed for mobile parts and low frequencies ... it can still be overclocked though to 3.9 or 4 GhZ fairly reliably ... and with some samples to 4.1 GhZ... but the process is obviously preventing further voltage and frequency increases as its not designed for that.

12nm LP according to the technical specs as released by GLOFO appears to be designed for high clocks and high frequencies... this should theoretically allow much higher speeds. But it also depends on whether that process can handle higher voltages (again, high power and performance can and usually implies that).
Obviously, Ryzen+ will be able to operate on lower voltages and stock frequencies than Ryzen 1 (probably lower voltage on higher frequencies as well)... we know this from the technical specs again as it mentioned lower voltages... but, we don't know how much higher in terms of voltages can it actually go as we only have a leak or two to base current info on.

If you observe the Ryzen 1700 in my GL702ZC laptop... technically, I can push it to 3.6GhZ at 1.175V (which is a notch lower voltage than stock). And 3.2GhZ across all cores can comfortably work with 0.975V or 0.98V (though obviously, OEM's don't really do this kind of voltage control to improve thermals and make the laptop more efficient - they provide lower voltages by default with later silicon releases which has better yields).

 

Well, that only proves that is is not an acceptable enthusiast product. 4GHz is pretty damned lame. You can manage a mickey mouse 400MHz bump to 4GHz from some of the most pathetic BGA filth, but it is not something any serious overclocking enthusiast is going to be impressed with. So, I agree with Brother @Papusan... it's a disaster for overclocking and a terrible disappointment to people that enjoy overclocking. It's a belly-button consumer product. That doesn't mean it is broken, only that it's nothing that warrants excitement because it's boring.

 

So?
Ryzen 1 was a 'worst case scenario' for AMD... they said so themselves... and whoever said that Ryzen was an overclocker dream?
I don't recall AMD saying that.
Plus, in a specific price bracket it was able to perform on par or outperform competing Intel products (especially when receiving software optimizations).
The only reason Intel has a clock advantage is due to their process being suited for higher clock speeds... but even overclocked to 4 Ghz, Ryzen manages to keep up mostly and draws less power than comparably clocked Intel system.

It's hardly boring either because it provides an alternative to Intel by bringing more cores to the table for a lower price (and it clearly outperforms Intel in a variety of areas in multicore - whereas Intel also stops benefiting from clock increases), clear upgrade path up to 2020 (which saves you more money down the line), and on the security front it's not susceptible to Meltdown at all, while having extremely low possibility of being exploited for Specre (which is fixable via simple OS update), and of course AMD's soldering and use of quality TIM vs Intel makes things a lot easier.

 

That just sounds like a list of excuses to me. AMD GPUs have sucked at overclocking for a long time, and now their CPUs do as well. They never said they were good at overclocking because they knew they were not. Since everyone has a belly button, they just aren't that interesting. Unless, of course, a person has a weird thing for belly buttons. Maybe they'll do better next time. I hope so, but we will have to wait and see how it turns out.

 

Not everything needs overclocking, the performance is what it is, and for MSRP it's a good deal.

Overclocking is untapped headroom, headroom that could be sold and profited from, that Intel keeps leaving it on the table is proof that they do sloppy work.

If AMD feeds the OC in the next or following CPU's, you'll find something else to complain about, which is fine, that's what we do,

Not OC'ing when AMD performs better $/$ against Intel /Nvidia makes OC'ing a non-issue, if you want more performance spend more money and waste it on Intel / Nvidia, just to have higher numbers.

The rest of us will get AMD CPU's / GPU's and be very happy.

Unfortunately, no GPU's are a good deal right now, or if there are good deals I haven't seen them.

 

Nothing will make me happier if AMD can deliver equal clock speed, equal IPC and equal OC headroom Perfeckt. If the price is lower as well... Yees, bring it to me But all boxes must be ticked. Especially the 3 first.

 

You keep missing the point. The point is AMD doesn't sell higher performance CPU's than it already offers - don't let that English confuse you, it's quite simple.

AMD offers their highest performance CPU's, while Intel offers higher core count and higher clock count CPU's for far more $$$$.

AMD's offerings stop at some point, and Intel offers higher core count and clock CPU's for far more money, $1000-$2000 on the consumer side.

Comparing the same cost CPU per CPU, if AMD offers what you need - which for 99% of us it does, then buy AMD.

If you want to spend huge amounts of money for the fastest CPU then waste your money buying Intel.

Bragging rights are expensive, I don't need them, and neither does 99% of just below highest end computer buyers.

The Ryzen 8 core / 6 core motherboards, CPU's, and matching GPU's (at MSRP), are far better buys for everyone.

ThreadRipper takes it up another notch.

If you have to blow even more money to make you happy buy Intel's overpriced benchmark winning CPU's, and get water cooling, and huge power supplies, and more fans with more radiators.

You make a lot of noise for someone that doesn't want what AMD has to offer, and I seriously doubt you'd be intrepid enough to buy AMD even if it offered a solution that OC'd / benchmarked higher than Intel - and perhaps even cost more than Intel's slower CPU.

So please stop polluting the discussions of people wanting to exchange information on AMD CPU's and GPU's, go bother someone else, your points are invalid for 99% of those of use that will buy AMD CPU's and GPU's.

And we don't care that you don't want to buy them. Really, we don't.

 

I will be fine of 2.5GHz APU if it delivers performance of 4.0 GHz Intel CPU.

 

Some of that makes sense, and some of it does not. Intel sells CPUs with untapped headroom because they can and because there are greater possibilities as thermal management improves. AMD sells CPUs that don't overclock well even with the best thermal management solutions. If they were to release CPUs that overclocked as well or better than Intel and had benchmark scores that beat overclocked Intel CPUs in a wider variety of benchmarks, more enthusiasts would buy them. Picking one or two benchmarks where they do better and ignoring the rest where they do not doesn't work for performance enthusiasts.

Nobody is criticizing AMD for selling broken products, and nobody is suggesting their products are broken. If a person is happy with really great stock performance and does not want to overclock, there's nothing wrong with that. They can be happy and save a few bucks by purchasing a product that runs well and sucks at overclocking. There is no reason for them to care that the 1% don't want to buy them. Nor should they be concerned that others might be swayed by facts presented from an opposing view. To suggest that it is bothersome to have an alternative view smells like a hidden agenda or a cover-up, but it can also be an emotional response to opposition.

This is only a personal opinion. It's not a dilemma unless the goals are for the bar to be set low and everyone to have identical opinions and expectations.

Those that are shopping for something new with an open mind deserve to know all the facts, and this is a forum where all of the facts warrant discussion. Same applies to the BGA turdbooks and AMD GPUs. Then they can make fully informed decisions rather than trust their "friends" on Facebook and carefully crafted marketing hype. Under the "go away son, you bother me" approach, our threads become the domain of sheeple, yes-men and Kool-Aid drinkers... kind of like Razer notebook forums. The hypothetical 1% that have higher expectations deserve to find the truth they seek. This is the best way to avoid disappointment and unmet expectations.

This is what I am waiting for as well. It's not here yet. Don't know if it will come later. We will find out when we find out. Until then, we have Intel to provide what we expect and will have to pay extra to have our expectations met.

 

TBH, of computer hardware enthusiasts only about 10% of them are overclocker's and about 5% of them or less are hardcore overclocker's. Just go see the percentages here, of the literal thousands of users there are only a few hardcore overclocker's. So I am sorry if I do not place much weight into what you say as a reason for AMD being a failure, it is more than good enough for 99%+ of the users out there and I am sorry they are not catering to you well under 1% of users.

My concern is stock clock offerings and performance. I would like to see AMD offer better to match Intel offerings. If I can get a bit more overclocking, all the better though.

 

It is nice to enjoy the pleasures of being part of an elite group. Absent that, I swiftly lose interest the the PC as a form of entertainment and pleasure, and it becomes a simple tool to do a simple task, much like using a measuring tape, hammer, or screwdriver, or making phone calls on my smart phone, my Roku to stream Netflix, Amazon Prime and Hulu, or taking my temperature with a thermometer. Those things only need to do what the manufacturer intended, nothing more or less, the same as most ordinary consumer-oriented computing devices.

Question: How are we coming up with numbers like 1%, 5% and 10%? Are those best guesses (which is totally fine if they are) or has an authority conducted an in-depth comprehensive survey of millions of self-proclaimed PC enthusiasts and gamers to capture reliable data? If there is such an authoritative source of data, it would be good to know what the real numbers are. They could explain the sorry state of things, or they could expose the fallacy of the marketing hype that seeks to manipulate the masses.

 

Laptops with mobile GPU's seem like the only affordable alternative (at least for the time being).
GL702ZC perhaps? Or waiting for 2500u/2700u with a Vega dGPU?

 

About the percentage of people that overclock - I reckon we could use the sales of 6700K / 7700K /8700K (the K-CPUs) to work out a rough percentage. I would think most people buying a K CPU will be overclocking it, as there's not much point to them otherwise. What's the market share of K CPUs?

EDIT: couldn't really find any figures on this. But if you look at the Steam hardware survey ( http://store.steampowered.com/hwsurvey/), and look at Intel CPUs, you'll see that 5% of Intel CPUs are 3.7GHz and higher, and all modern K CPUs since 6700K would fall into that bracket (my 6700K has a base frequency of 4GHz). This doesn't account for owners of 3770K/4770K, and older CPUs because their base frequency is less than 3.7Ghz. So, I reckon it's at least 5-10% of Intel CPUs sold and then found in use on Steam are K overclockable CPUs. Let's say 10% - then you could say 10% of gamers are overclockers, and if gamers (Steam installations) are 25% of the overall desktop CPU market, then you're looking at 10% of 25% which is 2.5% of CPUs are overclocked only. (It's probably a fair bit less than this I think, I was generous in the 25% figure and maybe some others)

 

Surprisingly, I run into more people with K CPU's that have no idea how to OC, or what OC'ing is. They get K CPU's in a lot of builds where people are just trying to get a good gaming computer.

I'd guess less than 10% of the K CPU's actually get OC'd in their original owner service life time.

I'm sure it varies greatly between desktops and laptops with K CPU's...

Most laptop buyers must seek out laptops that support K CPU's so I'd assume more like 50% get OC'd.

The percentage of HK laptop CPU's OC'd is probably closer to 10% like the desktops. Laptop buyers looking for the best gaming laptop would get an HK CPU just like the desktop buyers, but not be interested in tuning.

We actually see a very small percentage of users seeking out information on tuning to lower temps, and not all of those are interested in OC'ing either.

You and I couldn't imagine any situation where we wouldn't OC and tune for best performance / thermals if possible on our own hardware, most others don't even consider it.

 

(I've edited my previous post with some more info while you were typing your reply.)

I think it's more than 10% of K CPUs desktop owners that overclock, I think it would be more like 75% of them, I agree with those other figures you came up with though for laptops though. Laptops aren't as conducive to OC'ing.

 

To me its comparing apples to oranges, as Intel had been on basically the same process for how long? Saying Ryzen is a failure is unfair, and untrue. I have a stable 800mhz overclock on a first gen process. Saying it doesn't overclock as good as Intel is true, but wait for it to mature a little before throwing it under the bus. Just because I don't have a 13 or 1400mhz overclock doesn't make it any less fun, it just makes it harder to compete. I am in no way disappointed with my purchase, as I have a clear upgrade path for future Ryzen cpu's, and it's a monster multicore beast.

Sent from my SM-G935T using Tapatalk

 

Nobody is saying Ryzen is a failure. That would be unfair and inaccurate. The opposite is true. AMD did better than expected and should be commended for their effort and progress. Having a nonchalant attitude toward the fact that they have very limited overclocking potential is a failure.

We shouldn't make excuses and cut them any slack for it not doing well at overclocking as some would like to present as being the right approach. That only makes matters worse and gives them an out to not try harder next time. Saying it is OK that the overclock potential is poor because 99% of the people that buy it either don't care or don't know how to overclock is lame, and that is a very dangerous mentality that merely encourages more of the same kind of product to follow. To compete successfully with Intel (and NVIDIA) to take more of their business, and spur competition, their products need to compete at all levels, not just stock performance. It doesn't really take a special product to compete with soldered mobile processors and the locked down non-K/X Intel CPUs. Targeting that over-saturated consumer market segment that is content with low-cost status quo mediocrity removes them as a serious contender in the performance PC market.

That approach should never fly. It's not much different than saying that everyone should only buy Mustangs, Camaros and Barracudas with turbocharged 4- or 6-cylinder engines because they run better than the non-turbo wheezers, and because 99% of the people that buy them don't need the performance that a supercharged V8 can provide. It totally ignores the fact that some expect and demand more than others. Need and knowledge have nothing to do with it. It's all about desire and people having what they want, not what someone else thinks they ought to be content with. Some want something with potential for greater things even if they never tap that potential. If they don't bring it, then they're not a serious contender.

I am watching to see what they do next. Impressive first try. I expect more, and hope they close or surpass the overclocking gap. If they do, chances are really good that they will get some of my money. I need to see the results and Intel's response to that progress first though.

 

There was a time when CPU's were a bit more anemic and in the 1GHz range where overclocking mattered to me a great deal. Since graduating beyond C2D's this has become a non issue for me. I overclock where I can but I do not need the superclocks anymore for a great system.

Now prior to Ryzen using an AMD was like going back to a C2Q at best, not something I was willing to do.

 

Strange, that 99% of the people have missed the AMD deal Not all want trow out unnecessary amount of their hard earned money, then why is AMD's market shares lower than what they already is? What have AMD? 30% ? Or 40 or 50% ? Strange!!

FYI. None have said AMD is a failure.

 

The issue is not everyone needs a new system. The issue above that too is few systems are out there as OEM with Ryzen CPU's. Yet another issue is AMD systems out there on the4 cheap are OLD non Ryzen AMD CPU's. Yet another issue is even though the GPU may be only a 1050 in the new system Intel's gaming better sticks with consumers even if it is not true at that hardware level. Should we go on?

 

The ignorance of consumers has always been, and will continue to be, a source of grief for the small group that actually knows anything. The typical PC gamer is a converted console jockey, and those folk fall squarely into the misinformed consumer category. They might be serious about liking to play games and may be really good at it, but they usually don't know much about the hardware they are planning to use for doing that. I don't know the numbers and won't venture a guess, but it seems that there are a ton of them, (probably the majority,) that are not technically savvy and can't tell when they are being fed a load of baloney.

 

HP Envy 360 15z - Ryzen 5 2500u Vega 8 Graphics

Published on Feb 5, 2018
The HP Envy 360 15z is a 15.6" 2-1 laptop powered by AMD's Ryzen 5 2500u CPU with integrated Radeon Vega 8 graphics.
It has Windows Ink support using an Ntrig pen.
Base model can cost $650 direct from HP with 8GB RAM.
Other options available on Amazon:
http://amzn.to/2nJM9Rl
It promises to improve gaming performance without sacrificing too much CPU performance.
This is a full review

 

Ryzen 7 2700U (15w) Review (Acer Swift 3) vs Ryzen 5 2500U (25w) (HP Envy x360) vs i7-8550U (Acer Aspire E15)
Ahh, that's why his 2500u sometimes outperforms his 2700u, the power limit on the 2500u is 25w and the 2700u is 15w...described in this video.

*impressive* Tomb Raider 1080p high AMD Ryzen 7 2700U Vega 10. Acer Swift 3 Gameplay Benchmark. Test

PUBG "Adrenalin" Retest AMD Ryzen 7 2700U Vega 10. Acer Swift 3 Gameplay Benchmark. Test

Paladins AMD Ryzen 7 2700U Vega 10. Acer Swift 3 Gameplay Benchmark. Test

AMD Ryzen 7 2700U Vega 10 "Adrenalin" vs. Nvidia GeForce MX150 (Intel i7-8550U) in 7 Games

Middle-earth: Shadow Of War AMD Ryzen 7 2700U Vega 10. Acer Swift 3. Gameplay Benchmark

Deus Ex: Mankind Divided AMD Ryzen 7 2700U Vega 10. Acer Swift 3. Gameplay Benchmark Test

Subnautica AMD Ryzen 7 2700U Vega 10. Acer Swift 3. Gameplay Benchmark

Battlefield 1 Multiplayer 64 DirectX 11 & 12 Ryzen 7 2700U Vega 10. Acer Swift 3 Gameplay Test
https://www.youtube.com/watch?v=xhJJ5pR2z5Q
Street Fighter V Arcade Edition Ryzen 7 2700U Vega 10. Acer Swift 3 Gameplay Benchmark Test
https://www.youtube.com/watch?v=WidB6lr4p0g
Ultra Street Fighter IV 1080p max Ryzen 7 2700U Vega 10. Acer Swift 3. Gameplay Benchmark
https://www.youtube.com/watch?v=qY1xVKHLKvI

 

So, more than playable on 1080p and various high settings.
Now if the 2700u wasn't limited to 15w but instead allowed to go up to 25w as its supposed to, it would likely eclipse the 2500u and reach or surpass mx150