AMD vs. Intel vs. Nvidia: How Users Are Switching
boilingsteam.com
boilingsteam.com
Couldn't disagree more. For the last 4 to 5 months or so, Nvidia's drivers have been horrendous. I've experienced numerous crashes with my GTX 2080 Super card. About once every two days I experience a kernel panic and the only solution is to reboot. There are multiple threads on this issue on Nvidia's official Linux forum and Nvidia's driver developers have stated they don't know what the cause is and cannot reproduce it. [1]
My HDMI also doesn't work well after the computer wakes-up from sleep. Sometimes the monitor does work, sometimes it doesn't. It's also a known issue.
I'll be looking at AMD and their graphics cards a lot more closely when I build my next computer later this year. I just can't deal with Nvidia's driver issues. It's extremely annoying and it has wasted a ton of my time.
I've been using Nvidia's HW with GNU/Linux for about a decade and have never had any major issues and have always been happy with the performance and reliability. However, this is not the case anymore.
I'm pretty sure all this mess started when Kernel devs started blocking Nvidia's "GPL condom" effort [2].
[1] https://forums.developer.nvidia.com/c/gpu-unix-graphics/linu...
[2] https://www.phoronix.com/scan.php?page=news_item&px=Linux-Ke...*
Moreover, they stay silent in the forums now. They don't listen to customer feedback or at least give any signs of hope or assurance. Their company image became too dominant and arrogant (I think it runs more than skin-deep).
Some of the nVidia's huge market share is coming from the momentum they have. Linux users do not refresh their hardware too frequently generally.
I'm running a GTX680 on my primary Linux box, but it won't be nVidia next time. I'll be returning back to red team and I'd be a happy camper.
> I'm running a GTX680 on my primary Linux box, but it won't be nVidia next time.
Well, you're kind of making their point. Your business is probably insignificant for them, you have an 8 year old GPU :-)
Linux users don't have to refresh hardware as often. Long-term support is a cornerstone of linux. Linux gamers may spend less per year on gaming hardware but they are not broke. Give them a reason and they will drop that banked cash on anything they can trust to play nice with linux.
I'm well aware of that (since I'm one of them too), however I didn't make it clear I presume.
Thanks for highlighting it again.
It's also stems for efficient utilization of the hardware, too.
The flip side is that if you want to do GPGPU work (even if it's just local testing) you more-or-less need an nVidia card.
ATI was left in the dust while the foundation has been forming, unfortunately, but the history is not that straightforward.
TBH, while I appreciate what nVidia is doing, I don't appreciate how did they came to this point.
I have some close friends who're doing game engine and very low level 3D development. His thoughts about nVidia's driver behavior, even under Windows, is worth another rant.
Have they ever taken proper care of the visual part? I've thought that it exists in the first place because of the SFX industry's heavy use of GNU/Linux and Nvidia. That's a fraction of their market share so only got a fraction of their attention, likely mainly to avoid vacating the area for competitors than to actually make money (bad ROI compared to Windows).
Windows is essentially just the gaming market. Gamers are fickle, only buying a card every couple years and are very price conscious. But AI, and any other datacenter uses of GPUs (mining) buy cards in bulk, are not fickle, and are not obsessed with daily price fluctuations. Those are the markets Nvidea wants to be in and they are dominated by linux. Not playing nice with linux gamers will come back to haunt Nvidea.
Also, one should not underestimate the bias that gaming generates. All those linux gamers who suffer poor drivers while gaming saturday night walk into jobs at datacenters and AI research groups each monday. They remember. I certainly do. All things being equal, I lean towards AMD whenever the boss asks my opinion on our next purchase.
There are also Quadro cards aimed at professional OpenGL acceleration for GIS, 3D modeling, CAD, etc. These applications run on both Windows and Linux (soem are cross platform, some are platform locked) and, the ROI pie is a complicated picture.
Windows is dominated by gamers and casual users, that's right, however it's not the sole market for gaming. Steam is pushing for Linux in a relentless pace. Steam in Linux is getting rooted like a Sequoia tree. When it becomes visible, its roots will be much deeper and stronger than the trunk (which is a good thing).
nVidia used to care for Linux users. Extensions required by composition, off-screen rendering and other stuff came to nVidia first (when compared to AMD). Video acceleration is also the same (AMD's video acceleration technologies can were much better than nVidia due to their legacy but, they acted slow. They're even now. Maybe AMD is still better. I need to test).
This trend reared its ugly head ~3 years ago for me. I've upgraded my monitor to a DP one and DPMS went out of window. Screen suspension and standby has gone. They didn't fix it for older cards for ~1 year IIRC. Currently I'm experiencing a strange frame-rate drop. Will investigate when I have time. If an onboard Intel card can drive same DE (KDE) on a same resolution screen better with a much inferior CPU when compared to a triple-slot, out of the box overclocked GTX680, then there's something very, very wrong.
nVidia was top of the line, just works out of the box GPU vendor. Now due to its arrogance and dominance, cracks start to appear. Even Intel's driver blunders are solved faster (ah, they're open source).
In their defense, from their point of view, gaming on Linux is a tiny fraction of their overall business and, presumably, even their Linux one. For them, Linux is the same as HPC and ML model training and, maybe, cryptocurrency mining.
These companies aren't rewriting tens or hundreds of millions of lines of code for Linux. The difference is probably a fraction of a fraction of a percent.
Well, they don't — AMD drivers for new hardware famously take a few kernel releases to stabilize. (G)GP specifically noted their stability problem started only 5 months ago, which means at least that issue is not really systemic. The real differentiator here is handling of existing complaints: Linux has mailing lists, Mesa has a normal open for everyone bug tracker, while Nvidia has a user support forum, which they mostly ignore.
Closed source GPU drivers are not ordinary drivers. They have a lot of quirks and workarounds for programs, especially games.
When a dev studio makes some mistake in its engine or the engine doesn't fit well to current workflow of the driver, a new driver is shipped with specific workarounds for this game. These quirks generally add shims between the program and the driver by basically modifying calls to the driver to increase performance.
e.g.:
- A game is calling too many clears while it shouldn't? Pool them in the shim layer and let them through in batches of 1000.
- Is a particular benchmark is overloading the card by forcibly drawing off screen portions and hurting scores? Insert a shim to prevent drawing off-screen parts and chea... ehrm, optimize the performance.
- Is an important legacy application exploiting a quirk in an older driver for something? Introduce that quirk via another shim.
When you try to port these drivers as-is to another platforms, things get awry. Also the drivers were tremendously Windows specific. This is why FGLRX had (or does it still have) various daemons, HAL layers, during its teething years.
OTOH, free/OSS drivers has no shims, workarounds and other stuff. They are just drivers. Plain, simple, vanilla. They are easier to maintain and develop. Radeon driver is such a driver. Also AMD/ATI has some more tricks up their sleeve for more stable, better drivers regardless of their platforms.
- IP separation: All of the NDA bound parts of the GPU (currently it should be the HDCP stuff only) is a different IP module on the GPU. The cards are designed so that even the Radeon driver can work with HDCP module while being completely blob free. ATI/AMD specially altered silicon and GPU architecture to enable FOSS drivers on Linux and I think its effects are huge.
- "Factory Team:" Radeon driver is carried forward by volunteers but AMD also has a dedicated team working on Radeon driver full time. This group cannot communicate with FGLRX team to prevent tainting each-others wells. So, Radeon is blessed by AMD/ATI on hardware, software and management level.
When you publish your specs, dogfood it with another team, and optimize your GPU silicon and board to accommodate these open source exceptionally well, you have a killer combination.
The quirks part is a bit of a meme. Mesa contains a few application-specific overrides here and there: https://github.com/mesa3d/mesa/blob/c62996796cc46f1a1406f365..., https://github.com/mesa3d/mesa/blob/a678ec9b8c057311ed7e9697.... The reason Linux drivers, including the one from Nvidia, have fewer quirks is simply that Linux has fewer 3d applications overall. Those quirks also reside in userspace driver parts and should not affect system stability.
> When you try to port these drivers as-is to another platforms, things get awry.
Again, quirks are a property of a relatively high-level userspace part of the driver. There is nothing hazardous about porting them.
Quadro cards are used for real-time rendering of ultra-high resolution models for post-compute analysis or pre-prod/pre-analysis design work.
Quadro cards are not more useful in CUDA computations when compared to the Tesla cards.
They're different beasts for different scenarios.
Disclaimer: I'm an HPC administrator.
I would argue that open source can be more pragmatic these days. A lot of software offers the features people need. Managing a system based upon open source software is significantly easier since there is no reason to impose restrictions on use or redistribution. To choose an example relevant in this case, open source drivers can ship with a Linux distribution so you have functional drivers the moment you boot from the installation medium. In the case of proprietary drivers, there is typically an extra step once the installation is complete.
- Desktop: Linux vs. Windows - worse support for gaming (e.g. anti-cheat), no support for major creative software like Photoshop/Affinity or Ableton/FL Studio
- Mobile: FOSS Android distro vs. Google Android - no guarantee that apps I currently use will work without Gapps, no paid apps or in-app purchases
I have really been thinking a lot about this in the past months, because I would like to improve my privacy. The only real play seems to be moving to Apple's ecosystem, and buying a console for the games (and paying a lot more and losing customizability and repairability in the process).
Many music users today are locked into proprietary applications analogous to Studio Vision and their proprietary formats. If the company goes under, or if management decides to kill the product, there goes all your work.
Theoretically proprietary software can support open file formats — but there are such strong incentives for companies to leverage network effects and lock in users that at any moment you are one upgrade away from a policy change and silent conversion of your document to a proprietary format. Only FOSS guarantees that your documents will survive the tumult of capitalism.
Two decades later, history has repeated itself, with Flash, Macromedia, and Adobe in the roles previously played by Studio Vision, Opcode, and Gibson...
Partly this will be because I'm not playing games like CS:GO not cyperpunk, but I'm capped on 144hz with latest OSS drivers.
I've had to test out several combinations before I got a relatively stable config.
Now I'm afraid to apply an automatic update since I don't want to lose several hours to debugging BS driver issues.
That's where I was on my last rig (AMD+nvidea). Every driver update was a dice roll. Am I going to play today, or spend three hours googling to figure out why my machine keeps booting into safe mode? Since I switched to AMD+AMD a few months ago: no issues. Maybe nvidea would give me a few more FPS at the same price, but you cannot put a price on "it just works".
They are 'OK', I wouldn't say they're excellent. I have an RTX 3080 (prior to that I had a 1080Ti) and games crash more often than I would like. Waking from sleep is hit or miss, random driver crashes where the monitor goes blank, driver reinitializes itself, screen comes back on, etc. This is all with the latest and greatest drivers, 4K 160hz monitor (and an older 4k/60hz monitor), etc.
Turns out that it's the NVidia drivers every time.
So I jumped on AMD, missed the rtx train, but my card is still running strong in 2021 and I'm not in a hurry to switch; I'll wait until the release of a midrange ready tracing card from them.
Same with my trusty 1070. The WHQL driver from 2019 didn't crash, but it is now too old for current games.
GTFO with that BS.
My guess is the huge file size is due to the GeForce Experience tool that's included. It would be much better if it was a separate download, but you can choose not to install it during the setup process.
Imagine any other core component of modern systems requiring such extensive, closed source binary blobs?
The APU combos are particularly compelling for desktop Linux.
So on some of the later HWE Kernels 5.8 for me that's not so good just yet for getting OpenCL working on AMD's Ryzen/Vega integrated graphics(Ryzen 5 3550H) and Radeon RX560X discrete mobile Graphics as well on the same laptop.
So Blender 3D currently on my laptop can not detect any OpenCL/Cycles compatible rendering enabled Graphics/GPU and will default to Cycles rendering on the CPU cores instead(Way Slower and ties the CPU cores/threads up 100% until the CPU rendering is finished).
So the MESA OpenGL and Vulkan drivers are in good shape but really any OpenCL compute acceleration on AMD's Ryzen APUs and on Kernel 5.8/Later and that's a wait and see. And Blender 3D, Gimp, Dark Table and other open source Graphics applications that need OpenCL will have to wait for that for me.
So Compelling for gaming maybe yes but compelling for GPGPU compute workloads that rely on OpenCL not so much compelling on the newer Razen APU based hardware and AMD's Instructions and not as polished as some of the Intel/Linux related instructions.
But really most laptop OEMs are still not tuning their laptops firmware for Linux except system76/other linux laptop OEMs! But that ASUS TUF FX505DY laptop was on special sale pre pandemic for $499 so not a bad deal but with loads of special issues to surmount just to get a Linux Mint 20.0 Live USB to even boot to install Mint 20 in a dual boot configuration alongside Windows 10 10/1909 Home.
But Blobs and out of tree Wifi drivers and that laptop's getting a Samsung M.2/NVMe to replace the WD Black M.2/NVMe that shipped with the laptop and the Realtek WiFi card sits right under the M.2/NVMe SSD slot so that's getting a more Linux Kernel out of the Box friendly WiFi card with better features as well.
I have no clue why this is not included in distros by default though.
This has been in the works recently though, I've seen PRs for OpenCL 3.0 and OpenCL Image, not sure if they're in yet. With both Redhat and Microsoft being interested in working on this area now, I expect to see improvements soon.
Just curious, what would they actually change in the firmware to "tune a laptop for Linux"?
- the kernel driver (AMDGPU) - the user space driver (Mesa, AMDVLK, AMDGPU-PRO)
The lack of CL support has nothing to do with the kernel driver, that's something that gets implemented in the user space driver.
My advice, assuming you're on HN with a technical background, is to create your own bot for monitoring product availability on your store of choice. It's a pretty small investment of time, and it frees you up from the time spent watching a Discord or refreshing store pages manually. Set it up to poll at a reasonable frequency, set up some form of push notification, and find that it's not that hard to get ahead of the horde who are all relying on someone else to do the dirty work for them.
Signed,
Someone who got a launch day 3090 from the Nvidia store
On the other hand, I did get a 1080ti several months after launch during the crypto-boom of 2017 using the same approach. I have also stumbled upon lots of "we added more tickets to X event" type situations before they are announced on social media and disappear in an instant (think sold out music festivals and concerts).
Trust me, there are people sitting there refreshing the page more often than 1x per minute, and as far as I know they didn't run into any blocks either.
My goal is just to mimic me sitting at my laptop refreshing a page trying to buy something as soon as it becomes available, so that I can use my time for something better.
I believe nVidia once had a policy of deliberately crippling the double-precision performance of their regular consumer graphics cards, in an effort to push their high-end computation-friendly 'workstation' cards, but that's not quite the same thing.
Not a rock-solid source, but: https://www.techpowerup.com/forums/threads/why-include-fp64....
Anyway, there's blame to go around, and these days I'm more inclined to look at how poorly retailers handle it. There are few (or none) backorders or queuing, and sometimes easily gamed systems. Either way they get their money, but it's not a good look. (It is not the end of the world, but annoying and doubly so when so many of us are stuck inside. Certainly much worse problems to have, I know.)
There is a huge amount of CFAA or worse stuff that needs prosecuting - it isn't just bots placing orders, it's mass amounts of hacked machines and hacked phones being used as proxies and cookies stolen for a valid fingerprint/less CAPTCHAs, wire fraud and forgery of documentation being accepted (instead of RPKI/proper filtering)
Anyone that has a slightly popular (at all, really) will get shady requests asking to "install a SDK" or "buy your extension" which installs malware on your end-users devices. This is the slightly legal/grey area part. The other part is where AFRINIC recently revoked a /14 for fraud that was nearly entirely used for proxies and abuse by a 1 person US company, based on WHOIS remarks, obtained fraudulently.
Closed source blob, OpenGL 4.1 with video hardware acceleration.
Open source drivers, OpenGL 3.3 with a kind of working video hardware acceleration.
Windows drivers, still OpenGL 4.1 and DirectX 11.
So much for the benefit of open source drivers.
If it is a configuration error, Canonical has been making the error since it decided to replace fxgl drivers.
I'm a convert.
(Windows 10, where you can expect the latest hardware to work, is closer to a rolling Linux distro in that way. Meanwhile, point-release distros like Ubuntu seem more fit for purpose as server, not end-user, distros.)
Isn't there actually a worrying amount that do? Wifi for instance.
You might already be aware of this, but a less annoying way to test Safari compatibility is by using Epiphany (aka 'Web') as a substitute, since it's also based on Webkit. I've solved several Safari bugs that way.
https://github.com/foxlet/macOS-Simple-KVM
Thanks for the tip in regards to Epiphany, I am aware, but it doesn't help with testing stuff like ITP unfortunately.
I read breathless reviews of how the latest card performs so well even though it's just $x MSRP and the reality makes that seem ridiculous. It shouldn't take 3 or 4 years to ramp up production that much. Yes, demand has increased, but again they've had years to react to that trend. I'd like to be able to buy a mediocre card for less than $500.
I think you are mixing up a few cycles into one. There hasn't been a GPU "shortage" for multiple years. But there are certainly shortages at different times in the past few years.
Nvidia had to manipulate their Financial Numbers to cover up for excessive stock sitting in Channel due to Bitcoin crash. It took them 3 quarters ( and more ) to clear those inventory out. AMD had similar situation but their impact were much smaller due to less percentage using it for mining.
This teach GPU vendor an important lesson. ( Edit: Imagine the worst case scenerio AMD has a new product launch and has ample of stocks while you have three quarter worth of old products not moving. )
Since then they are much more conservative with respect to forecast and planning. It is FAR better to have enough demand waiting for GPU than it is for GPU sitting in channel quite literally begging for their Distributors to sell them.
This conservativeness has a knock-on effect on TSMC planning. At the start of pandemic everyone thought the economy could take a tank, which means forecast and estimation were even lower. But reality is all of a sudden PC sales and GPU sales are hitting new height due to more Stay at home and Gaming. Bitcoin is pushing into new height so casual miners are back.
You are taking two conservative supply decision ( Previous over Supply and Pandemic ) and two Demand reality ( PC / Gaming and Bitcoin ). That is Four level of differences. Catching up takes time, and it doesn't help when the whole industry, not just GPU, but everything from WiFi, 5G Modem, SoC, CPU, Amazon Server GPU, NPU are all in demand. Both Intel and AMD grew 10% despite Intel having lower ASP ( i.e Higher Unit Sales ) and were selling as many as they could. The Silicon demand are far greater than even the most optimistic analysis has predicted.
If you want evidence check out this link. It shows the the lifetime prices of a GTX 1060 card since its release in 2016. It had an MSRP of $250, yet for most of it's life since sometime in 2017 it's been around $320 or higher, more than 125% of the MSRP, and, when this particular model was last available, over $400. Even used its almost always been above $250, with an average used price over $350:
Why multiple years? It's unlikely the current shortage will take that long, and it only started at the end of last year. In Summer 2020 buying a gpu was not a problem. Then the RTX 3000 series launched with basically no cards actually being available, and all following releases were like that from both vendors.
But both the RTX 2000 series and the Radeon RX 5000 cards have been easily available back then (okay, iirc the USA had some supply issues, but that was domestic Covid and not production related), so there was no reason for them to ramp up production for now during the last 3 or 4 years.
> Even used its almost always been above $250, with an average used price over $350.
It's a high end 1060. Which in practice means almost nothing but that people will overpay for it, expecting better performance and quality than they'd get.
The $400 price when last available is a pricing artifact of the end of cards lifetime. When they stop being available, some vendors will ask ridiculous prices to basically scam those that do not know they could get a newer better model instead, or are enterprise customers that have to buy the exact same model as the one to be replaced because of bureaucratic stupidity.
Also, camelcamelcamel tracks Amazon prices. US Amazon is the most ridiculously overpriced hardware shop I've ever seen, with only occasional good offers in between. You can't take those prices of specific models serious.
So yes, this 1060 might have had a "shortage" after it stopped being produced and even 1060s in general did have one at one point in time, but at the same time GTX 1650 Super, the GTX 1660, the GTX 1660 Super or especially a RX 480/580 had been readily available (after the crypto bubble of course).
I do hope that this is a revival of the desktop PC!
It was my impression before BTC that ASICs weren't particularly frequent, since FPGAs existed for a lot of applications, but mining has the pressing need for absolute performance and tuning and iteration that it would accelerate that space.
In the last year bitcoin is over 200%. Litecoin is almost over 100% (in fact, since April it is over 200%). I have no horse in this race, but I am not sure I would call it "not a bubble".
I was under the impression that no one bought GPUs for mining anymore in general, as it's a waste of money compared to specialized hardware. During the last crypto bubble I noticed a lot of regular people building mining PCs, this time I noticed nothing of that sort. However, https://www.notebookcheck.net/New-financial-analysis-suggest... suggests I might have been wrong about that.
Only real reason to mention bitcoin is that basically all cryptocurrencies go up and down in price with bitcoin, so when bitcoin goes up in price, other cryptocurrencies tend to do so as well, incentivising mining all of them.
AMD CPU for obvious reasons(4950x while I wait for the opportunity for a 5950x) and an RTX 3070 OTW(haven't been able to get a 3080).
I have been mostly satisfied with AMD drivers, but I have not owned some of the problem cards(like the Radeon VII). I do like the latest Catalyst and Ryzen Master software.. But the new crop of NVidia cards crush the AMD cards on ray tracing performance, and DLSS 2.0 is voodoo I want for 4k gaming and VR. Would really, really have liked to go with an AMD 6800 XT to support the underdog but.. Those two features are clutch when shelling out nearly a grand for a video card.
I have a good feeling about AMD in general and for their next GPU architecture iteration. They are supposedly working on an answer to DLSS(details are thin and NVidia are ML giants though) and this was their first generation with hardware ray tracing support so lessons learned should bear fruit.
I'm definitely very excited for RDNA 3. I think AMD's GPU momentum is just starting, similar to the 1000s series.
Hopefully OneAPI just makes this whole conversation irrelevant.
But what creates inertia is my big expensive G-Sync screen which ties me to Nvidia. My monitor is the only expensive piece of hardware that survives between machines. And that's the genius of Nvidia's proprietary sync solution: So long as the cards are anywhere close, I'd of course choose the one that doesn't force me to switch screens. So far it has luckily not been an issue since AMD haven't upset Nvidia nearly in the way they upset intel. So Now I'm running a Ryzen and an RTX in the latest build.
They also just completed the acquisition of Mellanox, a company working on 300G NICs (I have a few of their ConnectX-2's in my home lab). It would be very interesting to have all the networking, CPU, and GPU in one place to see what kind of things would come of it (as proprietary as they may be...).
I remember seeing that correlation many years ago, also on the Windows side, and likewise most people with an AMD CPU also had an ATI GPU. The acquisition of ATI by AMD seemed quite natural after that.
(I personally have the Intel+Nvidia combination, although it's very old, and I don't use Linux as my main OS.)
I agree that NVIDIA QUADRO GPUs have an edge in stability.
The AMD FirePros are very nice cards and AMD actively works on improving the drivers. The RX 580 is also a really nice card.
NVIDIA dominates in the vGPU and compute side of things in the data center.
I for one would upgrade my 5+ year old desktop, but I simply can't buy the parts (and I refuse to pay 2x MSRP to scalpers).
Personally I did not think it would change that much gameplay-wise when I picked up my RX5700XT since there were so few games in the pipeline. I decided it was not worth the premium. I can still do most any game at 1440p and 100 fps or more without the headache. Why should I pay the premium for GPU features I will never use?
DLSS though has been amazing. I hope AMD manages to get close soon with a similar feature, it’s brilliant for 4K gaming, and all my displays are 4K in my house these days.
* https://www.gamingonlinux.com/index.php?module=statistics&vi...
* https://www.gamingonlinux.com/index.php?module=statistics&vi...
AMD usage is rising on Linux and for obvious reasons.
Yes I had to apply a workaround but this part was easy.
Basically that’s how you do it in a libvirt environment: https://wiki.archlinux.org/index.php/PCI_passthrough_via_OVM.... In qemu terms this translates, I _think_, to having "hv-vendor-id=Nvidia43Fixx,kvm=off" in your "-cpu" line.
So don’t know how to do that with esxi but all in all you need to change your cpu hardware vendor id to some random string (that may or may not need to be 12 chars long) and not explicitely expose the fact that you’re a VM.
People say AMD cards are better if you're on linux, but I disagree, because it doesn't look like AMD is coordinating their product releases with linux kernel releases at all.
https://cpu.userbenchmark.com/Compare/Intel-Core-i5-1135G7-v...
The Zen 3 and M1 chips are unfairly being compared even in recent articles to 10th gen Intel chips when 11th gen is widely available. Considering there is an actual node change at the 11th gen it’s clear from the benchmarks that Intel has not actually fallen behind AMD —- they are neck-and-neck.
Considering the lack of disparity in benchmarks, though I am a long-time AMD fan (see my tongue in cheek username), I still think Intel is the better choice at this time for laptops. The reason is that the ACPI and general driver situation for Zen 2 and 3 is very bad. Power management and hibernation are totally unstable on Linux and Windows. I like my Zen 2 laptop for the better multithread capacity (slightly faster compile times!) but the Intel laptop is likely to last much better given that it will not destroy its own battery. The AMD machine discharges in a day even when hibernating, and in a matter of hours when sleeping.
Intel still has a chance to catch up. It all hinges on how fast they can move to 7nm. And given the “chip shortage” (i.e. limited capacity and likely yield/production issues at TSMC), Intel is well positioned to continue to dominate the market for quite a while by dint of having its own supply chain.
You are still right that Intel mobile is still performance competitive if you don't care much about the core count gap.
If you do the opposite and compare Intel 11th gen to Zen2 the differences are stark:
https://www.cpubenchmark.net/compare/Intel-i5-1135G7-vs-AMD-...
AMD is ~15% faster on the low-end and ~50% on the high-end. But as far as I can tell the actual comparison is Zen 2 with 10th gen in the same generation of laptops. For the Lenovo T14 for example this is the comparison between the high-end of Intel and the two AMD options:
https://www.cpubenchmark.net/compare/Intel-i7-10610U-vs-AMD-...
It's ~80% faster to ~115% faster. Zen 3 will probably maintain a similar gap when the next generation comes out. The 11th gen Intel processors look fine but particularly at the high-end the 4 vs 8 core difference is huge.
That isn't a CPU-related issue right? The CPU would be completely powered down when hibernating
For a desktop Ryzen sounds great. But for a laptop it has given me only frustration.
> On Linux, most AMD enthusiasts have probably started by adopting CPUs for quite a while since they fared better for parallel workloads – and now AMD is even catching up on single-thread performance with their latest processors.
I did the opposite, and started with AMD gpus, as those provided best bang for the buck in ~2008. Driver situation on Linux was quite dire, but it improved tremendously around 2013 (when Steam appeared on Linux).
Even though I was still advising AMD in the bulldozer era, Intel was winning performance-wise. On the GPU side, though, GPU drivers have been quite hassle-free since 2015, provided your GPU wasn't the latest and greatest. They got better at supporting their latest GPUs recently.
Intel and Nvidia always had despicable market segmentation and anti-consumer practices, which is the main reason I always went ATI/AMD.
Intel: No ECC on consumer CPUs, reduced number of PCI lanes, restricted frequency multiplier.
Nvidia: Signed GPU firmware, non-redistribuable. Driver limiting the use of GPUs in datacenters. Proprietary tech everywhere: G-sync, 3D vision, Physx. Artificially restricts the speed of nvidia libraries on non-nvidia hardware (Intel did that too, to a lesser extent). Cripples FP16 and FP64 on "consumer" hardware.
I recently got myself a HTC vive. It works fine, while not perfectly, on my R9 Fury under Linux. Go team red!
AMD can just work, but most people who have problems with new hardware are the AMD users. It comes mostly down to software support, drivers and BIOS configuration.
Another example is that every 'fps performance guide' always included a big section just for AMD users. It's just too finicky.. Always has been.
This still makes me shy away from AMD and just spend the extra few bucks.
AMD's sweet spot are the people on lowish budgets
What?? Actually, it's quite contrary. RTX has significant impact on the image in this game.
So, in lower resolutions/level of details AMD wins. In 4k Ultra NVidia leads by 4/5 fps. But you for 4k you probably want 6900XT/RTX3090...
Maybe a comparison of still images is not the best tool for “measuring” RTX impact.
Are people really buying new mid/top-tier graphics cards to play at low resolutions and levels of detail though?