The Loyalty to AMD’s GPU Product Among AMD CPU Buyers Is Decreasing
blog.parsec.tv
blog.parsec.tv
Here's to hoping the new AMD CPU does well too, it's good for the market that Intel has competition.
Also, as far as I understand the open source driver is only included with Linux Kernels 4.7+ while Ubuntu 16.04 is on 4.4 (16.10 is on 4.8). So if you want to use the open source driver you have to pick your distribution carefully.
If you like reinstalling every few months, the non-LTS release is fine. Most people don't, so they prefer the LTS release.
They release a LTS version every two years in April: 12.04, 14.04, 16.04 and the next will be in 2018, 18.04. They do add a "patch" version after that, but the YY.MM are the major releases. 16.04 is an older version than 16.10, but it will be supported with patches for a longer period of time (major versions of individual software packages will stay the same).
I have a bad experience with this, especially upgrading one LTS to another LTS. Clean install was much more reliable.
The 12.04 o 14.04 upgrade was also broken. I don't remember details, it booted, but the dependencies went weird. I wasn't trying to fix it anymore, just straight reinstalled from scratch.
I haven't upgraded it to 16.04 yet. I will try, what will happen, once Kodi stops supporting 14.04.
Second, if you are using PPA's you are much more likely to have problems. There's no way to test all the inter-relations between packages that are outside the central archive. If you have packages that have upgraded fundamental packages in Ubuntu's Main archive then it's likely to break.
I mention this because you're talking about multi-media.
In general, it's best to remove PPA software before you launch the upgrade, in fact it's a good opportunity to remove anything from the system that you don't really need: for example for my 16.04 upgrade I removed the C++ environment as I'm not using it.
If you look to the GPU market where AMD is in better shape, they do fuse off features like FP64 and ECC unless you pay for the heavily marked up FirePro versions.
With the push to the linux kernel AMD GPU's would be better than Nvidia... that being said, I wouldn't hold my breath.
If you want to write a kernel driver that's effectively a Russian train toilet nobody will stop you, and you can feel free to maintain the engineering effort on it - but it's not going to be accepted into the mainline Kernel and you'll have to maintain it out of tree.
If AMD wants a shim to make it easier to port their Windows drivers over that's perfectly acceptable, but they need to work with the existing DRM infrastructure and the people that maintain it to get a kosher driver that everyone can be happy with.
It only interacts with performance in that driving displays requires memory bandwidth and consumes power.
That said, they were warned 6 months before the rejection that abstraction layers in the kernel would probably get rejected. The rejection wasn't news in any sense - Dave Airlie just decided to be more direct about it because the warning from half a year ago was ignored. No one knew that was what they were doing because AMD did all the development in private and then just handed over the code. Had they been more open and engaging with the kernel developers, they could have saved a lot of time.
There were a few emotional responses, and then it was made clear that they weren't done yet, and were still willing to work together. So everybody calmed down and got back to work. This all happened within a few days of the initial rejection, so you really shouldn't push the "it's never going in the kernel" narrative. Because much of it will probably be in there by the middle of next year.
Still, comparing NVidia's current lineup to AMD isn't apples to oranges. If you want a 1070 or 1080 just get one, Vega is still months away and while we see AMD stomping in Vulkan and DX12 benchmarks compared against similarly priced cards (RX480 vs GTX 1060) they don't have a proper answer for the higher end cards until Vega is out.
Otherwise, don't let driver support deter you, the whole design behind AMDGPU-PRO is much like NVidia's proprietary drivers - it lets AMD reuse the majority of their Windows drivers to keep parity between platforms; and, again, the open source drivers only keep getting better at a rapid pace.
amd is a bit lower, but you can use the built in open source driver and get roughly the same performance as the proprietary driver (From what i understand, the proprietary driver is going to only be there to support their direct clients, and in general, they are working on getting all generic improvements into the open driver).
From what i understand, this means kernel updates mean you have to re-build the nvidia drivers, whereas with AMD, the new drivers are included in the new kernel.
either way, I went with an AMD rx480 this month because i dislike nvidia's closed off business practices (especially in comparison to AMD's open practices). it runs everything I want it to run just fine, so its not like i suddenly cant play some game because i went AMD.
I always wanted to try it, but on the flip side the fact that AMD was starting their own project to do the same thing doesn't fill me with confidence that GPU Ocelot is worth my time.
The difference is in the scaffolding (CUDA is easier but less flexible), and pre-existing libraries on offer.
If you're going to be writing the kernels yourself then it is easy to maintain both versions and it if it is for learning it doesn't matter at all (but I prefer the standard that works on all my machines).
Well, OpenCL 2.1 does do some C++ (not as much as CuDA with Pascal), but isn't widely available anyway.
If you were referring to the kernel language (a restricted subset of C99), I don't find that to be a disadvantage in practice. While CUDA supports a fairly useful subset of C++ (most notably, templates), I still find host-side metaprogramming to generate kernel code to be a much more useful approach in most cases. Of course your preferences may vary, but I haven't seen many experienced GPGPU people express dissatisfaction with this point.
I have no idea about the other languages but surely other bindings must exist?
> I have no idea about the other languages but surely other bindings must exist?
Yeah, the usual poor man's solution of generating C code, thus introducing two compilation steps, without any debugger support.
[1] https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP
I'm glad they're increasingly open, and it's good to see the excellent work happening in the radeon drivers et al. I've just been burned enough by them that my aversion has become quite deep seated. That and so many things favour CUDA instead of OpenCL for acceleration, which is incredibly annoying (especially as it's possible to leverage even inbuilt Intel GFX for that in addition to the main GFX card)
You can only believe the same promises up to a point.
My last rig was all AMD, and while it was really powerful for the time it was built, it was basically a space-heater. The power supply on my current machine wouldn't even run my old computer at idle. I have a philosophical issue with the way that Nvidia and Intel handle their Linux drivers and interact with the open-source community, but after 7 years with a computer that sounded like a jet engine, I just wanted silence.
I know it seems disingenuous to compare a 7 year old computer to a modern one, but the TDP of AMD products has not improved much during that time. Plus, my computer spends most of it's time near-idle, and Intel and Nvidia look better the lower the utilization. The fans in my new rig don't even spin up unless I'm doing something rather resource intensive, like playing game.
So, despite my misgivings about Intel, there was no chance I was going to with AMD. The Bulldozer architecture is terrible. Bulldozer hasterrible performance, regardless of whether you measured actual performance, or performance per watt. The only reason I considered AMD was because they enable virtualization on all of their processors (or at least the ones I was considering), whereas Intel only allows virtualization on Xeon and K-series processors.
My biggest issue right now is that Nvidia takes steps to prevent their non-Quadro cards from working in VFIO passthrough. I could have just as easily gone AMD, which I understand works much better with VFIO; the main reason I went with Nvidia for graphics was because I got the 970 for a song. Plus, after years of using AMD with Linux, I don't really trust AMD drivers to work. Nvidia drivers might be closed-source, but they work as long I wait a week to upgrade after a new driver is released.
The Zen architecture seems like it will hit all the right notes for me, as long as AMD maintains their welcoming approach to virtualization.
I'm not an expert on Intel CPU market segmentation, but I do work for Intel on the i965 Mesa driver. I don't think your statement is true.
See http://ark.intel.com/search/advanced?s=t&VTX=true
That's a list of all Intel CPUs with VT-x. It's really easy to narrow down the selection from there. FWIW, I see lots of Skylake i3, i5, and i7s with VT-x.
I'm not sure which modern CPUs and motherboards support that. I know back in the Core2 days when VT-d was first introduced, it was a nightmare to find the right combination. Looking from the outside, Intel also seems to play a lot of marketing games with their CPU features. Sometimes top-of-the-line CPUs will be missing features that others have. Like selling a $6K Xeon CPU manufactured this quarter without virtualization support. (http://ark.intel.com/products/95831/Intel-Xeon-Phi-Processor...) Supposedly they want to get that feature in there at a later date, but that's not something we like to trust in with computer components.
Your VT-x search returns 1459 products, while the VT-d search returned only 788. Here's a list of things that say they support VT-x but not VT-d (http://ark.intel.com/search/advanced?s=t&VTX=true&VTD=false). Pretty sure some entries on it are wrong, but Intel is also terrible about keeping old technology names on newer spec sheets. Maybe they folded VT-d into VT-x without telling the world?
I agree that this has been confusing in the past (as evidenced by the mix of Atom, Celeron, Pentium, i3, i5, i7, Xeon in your link), but I think at least in VT-* it's been greatly simplified.
http://www.tomshardware.com/answers/id-2031324/4770k-virtual...
From April 2014: Craziest thing. The i7-4770K NOW does support vt-d. It didn't used to, but went on their site today....
I checked the edit history for the Haswell architecture on Wikipedia, and it appears that VT-d was supported by all i5s and i7s, except the K series, save for the 4790k and it's i5 counterpart since at least August of 2014. I'm not sure where I got the idea is was more limited. I had been speccing a computer for a couple of years at that point, so I wouldn't be surprised if I'm digging up irrelevant information.
Without Chevrolet we'd be driving Model-T's. Without AMD we'd by running 8080's. Support the little guy, even though they've historically not always made the right choices.
I bought an AMD RX 480 this year. I hadn't bought a discrete GPU in decade. I looked at Nvidia, but saw you needed Quadro/Grid to use with VT-d. AMD's GPU work with VT-d out of the box.
It'd be a damn shame to lose a company like AMD that doesn't disable features for marketing reasons. I'll happily buy another AMD CPU if the zen line comes close to the hype.
The electricity cost will in itself mean in maybe a year or so they'll cost more than an equivalent Nvidia/intel card.
Also not sure how the hotter-running AMDs affect their life, but it's possible it's considerable...
Being honest, AMD disappointed me, it was my first AMD GPU, that I purchased after all the pain of nVidia Optimus, and... AMD managed to outdo nVidia in how shitty they are.
* All cards and models have terrible power usage and heat.
* Drivers, both on Windows and Linux still aren't decently stable, on my Windows I had to switch drivers SEVERAL times depending on what game I wanted to play, because with each game it had a different serious bug.
* AMD software (not just drivers) crash a lot on my machine. No other software behave like that.
* AMD tried to "pretend" they don't have too much power usage, on the 380X case they just put a tiny TDP limit for a beastly GPU, so it all the time keeps throttling due to power limits (even when increasing them... I am looking now for information on how to edit the TDP beyond the card defaults), the 380X TDP is the same as the 380, despite it having double the RAM and having more transistors to power.
* AMD RX480 "tdp cheat" instead was to add only a single cable, pretend the card didn't had too much power usage, and let it melt people's PCI-e slots by pulling 7.7A from slots rated to 5.5A
* AMD distribution network is terrible, they make zero effort to sell around the world, meanwhile nVidia and Intel trounces them, not only in marketing, but by reaching local companies to do distribution deals, for example I paid for my 380X the same price as a GeForce 970... (this was one month before RX480 launch). When nVidia launched the 1080 they called the local media, and told them what price the retailers were supposed to sell the 1080 (even if they ran out of stock), a price that was slightly cheaper than AMD Fury cards...
* AMD and partners support just suck, I asked the SIZE of my card, and their support instead told me to "RMA" it, I tried to explain I wanted information, not an RMA, and they refused to help... when I asked about the TDP, then thigns got worse, they got even more staunch that I should just return the card (and eat the shipping costs to US myself!)
* AMD official forums has employees spouting bullshit (like claiming in a huge thread of people complainign about the 380X, that it was the hardware on the complainers machines that was defective, withotu realizing that he just accidentally painted the whole 380X product line as shitty, since there is lots of people complaining of common issues, and if lots of people have the same issues, and is a hardware issue, then the hardware that is crap... also, on the forums making ANY negative comment about AMD, get you attacked, people claim you are nVidia shill or worse, I even got banned from AMD chat after I asked how to circunvent a driver bug that was preventing me from setting my CRT resolution correctly, because they didn't wanted me talking in public about negative things).
https://www.custompcreview.com/news/amd-aims-reduce-rx480-po...
The driver do two things to "fix" the problem:
1. Draw more power from the cable (it is what the card should have done in first place).
2. They put back on the card the behaviour the 380X have, that I am trying to get rid of, of clamping down hard on the TDP and throttling heavily.
And yes, the claims of this making the card faster ARE true, and also applies to older cards, AMD cards draw so much power and make so much heat, that if you UNDERCLOCK them, they can get faster, because sometimes the stable speeds with less throttling is a bigger benefit than the amount that you underclocked the card.
That said, last I checked (this was 2 months ago), the driver changes were Windows only, some guys mining bitcoin on Linux didn't noticed and happily melted their riser cables.
I really don't like nVidia behaviour as company, pulling shady tactics left and right, but AMD frankly shocked me with their shoddy engineering. (also see: recently Linux kernel devs REFUSED AMDGPU driver patch... AMD was warning in February that the patch they were attempting to do was crap, and they still went ahead with it, there was even some passive-agressive personal insults exchanged on official Linux kernel mailing list).
Of course they seem to have been correct that AMD was way underpriced where it was before - but at some point it'll come back down too.
The new architecture may change things, but for now, that's it. They might have started losing market due to shady tactics from the competitors, but the nail in the coffin was the botched bulldozer. that had consequences felt on every series down the road, including unrealistic power consumption per unit.
I ended up going with Intel, but if I had decided to go with AMD, I would have spent a while figuring out which CPU to go with.
I hope AMD simplifies its product line when Zen comes out and gets rid of the confusing nomenclature and the low-end crap that clutters everything up.
When I need a new CPU or a new GPU, I just open the latest benchmark and go for the best price/performance ratio for gaming.
Perhaps most AMD GPU products have seen a higher price raise in the past six months in preparation for this shopping frenzy?
The GTX 1070, GTX 1080, Titan XP, Quadro P6000, and Tesla P100 are all faster than every AMD GPU in similar price brackets.
Additionally, GloFo 14nm (used by AMD Polaris) is generally thought to be less efficient than TSMC 16nm FF+ (used by Nvidia Pascal).
I will continue to buying AMD CPU but I've already switched to Nvidia for GPU. I will switch back if the next offering is something like Maxwell cards, silent, low power consumping GPUs.
I always want them to do well, since I don't want just on company running the x86 market.
The FM2-based products are alright, but they are glorified laptop processors and they do not really compete well in the desktop market overall. Nice if you want a decent iGPU but most people use discrete GPUs for any serious gaming, and without the iGPU all you have is a mediocre CPU.
AMD's future in the CPU market rests heavily with Zen. Right now they essentially do not compete for the vast majority of users (power-sensitive mobile/server market, productivity users, or gaming). They run the games, the averages are sometimes decent, but the minimum frametimes suffer pretty badly and they use a lot of power. Compare the 99th-percentile frametimes and the cumulative frametime charts here (both are "badness" metric for measuring stutter) and you can see that AMD's single-threaded performance really torpedoes some games far beyond expectations. The FX-8350's 99th-percentile frametimes are significantly worse than a Pentium G2130 and it even falls behind a dual-core Clarkdale (first-gen Core i5).
http://techreport.com/review/23750/amd-fx-8350-processor-rev...
AMD's GPU products, on the other hand, are still reasonably competitive - although their top product only competes with a GTX 1070 and is very low on VRAM capacity during a time when VRAM consumption is increasing rapidly and will continue to do so (particularly DX12 and Vulkan games). The RX 480 is a solid card though, especially for the price, and the Fury is due for a refresh soon with the new Vega series, which will undoubtedly have more VRAM.
In particular, there's a problem with combining AMD GPUs and AMD CPUs. AMD's GPU driver stack has a reputation for being single-threaded and somewhat inefficient - so you really need good single-threaded performance with AMD GPUs more than ever before, and Intel processors are very much the preferred pairing. Again, this difference is particularly pronounced when comparing minimum frametimes rather than averages.
They have been working hard on cleaning this up with Crimson and they made another big driver refresh recently too - but AFAIK there's still a pretty significant quality-of-life improvement from using Intel processors with your AMD GPU due to minimum frametime improvements.
Conversely NVIDIA's driver stack has a reputation for being less dependent on good single-threaded CPU performance. So perversely, if you are running on an AMD CPU then you are best off getting an NVIDIA GPU.
Also, side note: AM1 is my favorite AMD CPU platform right now by far. The CPU supports ECC, most motherboards don't but the Asus AM1M-A does. It makes a nice little NAS box if you can forgive its paltry 2 onboard SATA channels and mATX footprint, and you can pick up a CPU+mobo for $45 from MicroCenter.
Maybe there was a pattern of brand-loyal, sufficiently rich gamers who made a hobby out of building a new system at the very top end of AMDs offerings whenever a new generation Radeons came out. Those would not find an excuse for an upgrade this year because the latest generation of Radeons tops out lower than previous generations.
But this explanation attempt breaks down completely when I try to back it up with the numbers from the article. The way I understand them, the average AMD/AMD system has actually become more expensive, suggesting that AMD lost more at the low end. But I have similar problems with the main conclusion of the article: what I see in those tables is that the Radeon fraction has dropped faster on Intel machines than on AMD CPUs. With this in mind I would rather conclude that brand loyalty is even more important for AMD than it used to be, it's just that with shrinking general popularity of Radeons, the CPUs get less of a helping hand.
This is a significant difference, because Intel has a 17.48% marketshare on graphics per the Steam Hardware Survey [1] (the original source for the spreadsheet data), while AMD has a 23.46% share -- which includes both their integrated and discrete graphics.
The average price of AMD builds has risen because the lowest-end builds on the Intel side can rely on Intel's integrated graphics. Intel integrated GPUs attain one or two percent share of all GPUs for each generation, while AMD APUs are far behind.
Everyone I know with an AMD GPU has an Intel CPU (Core i5/7) since buying AMD GPU's for gaming is an utter waste of money.
The CPU's are slower, and the platform is old and buggy, not modern PCIE, storage performance is garbage, and USB 3/3.1 support is laughable.