HN Theater @HNTheaterMonth

I'm not a fan of Intel, but honestly this was pretty well expected of them, the delays and driver issues.

And yes I know it's marketing, but if what they say in their LTT[1] video is true, then GPU driver optimizations are apparently a titanically tedious clustertruck, which... honestly, considering all the weird API's and graphical schenanigans of games past, I concur.

I think if they didn't release now, even with all the bugs, they were never going to. If it actually makes a profit, they'll finally have in incentive to invest enough that it even COULD catch up.

[1]https://youtu.be/45n5pnEyw9o

⬐ Sakos

NVidia and AMD drivers are 20+ years of piling hacks upon hacks just to make sure games run properly and a paper over obscure bugs. It's gonna take time for Intel to catch up.

⬐ ahartmetz

The funny thing about that optimization clusterfuck is that AMD's FOSS drivers get very close to the proprietary drivers in performance without any hacks.

Instead of millions of lines of code, the user space drivers (where optimizations happen) are more like 100k per major GPU generation (which can span like 10 years) + a few 100k shared between all Mesa drivers.

⬐ rasz

>close to the proprietary drivers in performance

name some examples, few games that get better fps and frame latency under linux+foss driver vs windows+blob

⬐ burmanm

AMD's FOSS drivers for Windows gaming? Where? As the article said, the real problem is in the gaming optimizations on older APIs.

You need to optimize the driver for each game basically, thus every supported game adds more and more lines in that silly world. This is for all the old DirectX9-11 games.

⬐ ahartmetz

That was about OpenGL for Linux. There are a few OpenGL using games that run on Linux and Windows, so direct comparisons are possible.

⬐ faeriechangling

>but if what they say in their LTT[1] video is true, then GPU driver optimizations are apparently a titanically tedious clustertruck

That would only imply that Intel's position is even more hopeless than it seems at first blush.

⬐ tpmx

IDK; I think they just need to abandon their core principles of only persuing something new for 18 months max. They do have the capital. FFS.

⬐ bilegeek

No company can change like that without being entirely gutted and remade.

...though they can have splitoff "cells", like the 5150 team or Xerox PARC. I can't tell from here, but given Intel's waffling in the past, maybe this team is a cell.

⬐ tpmx

maybe this team is a cell

let's hope so!

⬐ Hamcha

I'm glad Intel has been swallowing their pride on this one, I want them to succeed for once.

What really would drive me to support them directly would be (para)virtualization support via SR-IOV/GVT-g. Especially given Microsoft's recent stunts an increasing amount of people (including me and close friends) are evaluating moving to Linux while keeping a Windows VM around, but both NVIDIA and AMD seem adamant about keeping those features gated to enterprise cards (even thought it's a software thing, see community efforts to reverse this[1]) so they can charge ludicrous prices to businesses for those.

My opinion is that if Intel wants to get their foot in the door they should win over enthusiast by breaking patterns such as this one rather than try to compete exclusively on performance.

[1]: https://github.com/DualCoder/vgpu_unlock

⬐ preisschild

Their GPUs should also work really well in linux, since you don't really need DX11/DX9. All games with those graphic APIs get translated to Vulkan by [DXVK](https://github.com/doitsujin/dxvk) anyways.

⬐ Hamcha

You don't really need DX10 and lower on Windows anymore either, dgVoodoo2[1] is routinely used (along DXVK) on Windows itself because of how better supported DX11/DX12 and Vulkan are by modern drivers.

I hope someone will benchmark a game with DX9+dgVoodoo2 vs native DX9.

[1]: http://dege.freeweb.hu/dgVoodoo2/

⬐ easton

Off-topic, but if you're wanting to move to Linux and are worried about gaming, you should try running Windows games on Linux with Valve's Proton. I was worried too, but I got a Steam Deck and it's absolutely crazy how many games just work. I have generally mainstream tastes and over half of my Steam library is at least rated "playable". And there's rarely any tinkering you have to do beside click play (that seems to be one of Valve's sticking points for marking things playable, as if you had to do extra setup outside Steam it would be difficult to do on the Deck with no keyboard attached).

Most of the ones that don't are because of anti-cheat, which wouldn't work inside a VM anyway.

(On-topic, I want a consumer level card with SR-IOV too. I should be able to have my desktop be a VM in a closet without buying a license and card that cost more than a car.)

⬐ kllrnohj

> What really would drive me to support them directly would be (para)virtualization support via SR-IOV/GVT-g.

Linus asked the Intel reps for that on the WAN show and the answer was that Intel was also going to have that he an enterprise only feature at least initially but they'd try to escalate.

But I wouldn't get your hopes up. I'd love to have it and it's really dumb it's so locked up, but nobody seems to want to risk devaluing the enterprise markup.

⬐ carlhjerpe

I tried to hold out for the Intel launch anticipating sr-iov support, but I ended up refreshing hardware before. Any half-decent motherboard has their iommu groups proper, allowing passing any pci device into a vm with minimal overhead.

It sure sucks to need two cards, but I went/want AMD for Linux and still(sadly) NVIDIA for Windows.

It works really great. A 3 display setup where left is always Linux, right is always windows (when it's running) and middle is switched between inputs with ddcutil (cli tool to reconfigure your display).

I tried looking-glass framebuffer copy over a shared memory bus but it wasn't quite as smooth for me as switching inputs.

I'm on low-end gear, AMD 5600x, AMD 390, NVIDIA 3060, ASUS X570-P board. No tuning whatsoever, I let the Linux scheduler deal with it and it's good enough.

I can't recommend looking-glass enough as your spice client for mouse and keyboard capture, it's great.

Though this can all go fck right off if someone(Intel) does SR-IOV on consumer gear, I will replace my cards in a beat.

⬐ tjoff

Problem is, there is no decent AMD card (and I will not touch nvidia for linux) with decent number of outputs for that use-case (in my opinion).

The cheapest ones are like 8 years old and out of support and still costs like $250, comes with a fan and still doesn't have decent outputs. And also takes a lot of PCI-E slots. A dual graphics card system means you barely have expandability options left after you have the hardware required to boot. And that ~one PCI-E slot still visible might be disabled if you decide to use another NVME-drive.

The amount of research required to pull it off is insane.

This is my biggest hope for Intel/ARC. A low powered card with decent number of outputs for a fair price and decent linux support.

⬐ kllrnohj

What exactly are you looking for? There's both the single slot 6400 with dual outputs or the dual slot 6600 with 4 outputs. And both are modern RDNA2 architectures.

⬐ tjoff

3+ digital outputs, preferably passive and preferably one slot (though many compromises have to be made).

6600 costs about $350 where I live, which is quite steep considering it's only purpose is to serve 2D and very basic 3D editing. Especially considering you'd also need a more capable/expensive card for the VM (at least for my usecase) .

Compared to A380 with 4 digital outputs which costs less than $200, it has the potential to be an extremely attractive linux card.

⬐ kllrnohj

Have you considered a DisplayPort hub? MST support has been a thing for a long time, it should work pretty well (particularly since this is basically how ~every thunderbolt & usbc dock works)

⬐ tjoff

I have MST on my displays and it is a disaster. Also, considering the usecase where you use the VM to drive one of the displays which breaks the MST chain.

For this reason and how displayport/gfx-vendors handle unplugged cables I do prefer HDMI/DVI. Wish it wasn't the case.

Maybe a MST-hub could work well but I've heard of some of the difficulties and pecularities a simple KVM-switch can suffer from so I don't have high hopes for it.

⬐ acomjean

Gamers nexus had a few video reviews as well.

A a380 from China benchmarked.

https://youtu.be/La-dcK4h4ZU

Intel sent some engineers over after to explain the models and talk about drivers and tech.. https://youtu.be/AN8ZAf15DrM

⬐ fnands

Great to see them sending over an engineer for this one. Speaks to a kind of openness that's refreshing to see coming from Intel.

⬐ 2Gkashmiri

Has anyone talked about Linux compatibility of these GPUs ?

⬐ geerlingguy

I believe in a WAN show appearance last Friday, one of the Intel engineers mentioned it was in the pipeline, and I believe it was hinted they'd be more along the lines of AMD than Nvidia in terms of the open sourcedness.

⬐ TingPing

It's going to be supported upstream Linux/Mesa but nobody has tested/reported on it publicly. I'd expect growing pains but good support.

⬐ bobdvb

Yup, the code is already in: https://www.phoronix.com/scan.php?page=news_item&px=Intel-DG...

https://www.phoronix.com/scan.php?page=news_item&px=Linux-5....

https://www.phoronix.com/scan.php?page=news_item&px=Intel-Li...

⬐ ece

I've never had any problems with Intel drivers on their current Xe SOCs (ie. i7-1165G7), so the discreet GPUs with Xe should be pretty good I would think.

⬐ Havoc

Would make strategic sense to be fully open. Even if it can’t 100% compete on performance this would give them an edge

⬐ mfrw

off topic - cool username :) @2Gkashmiri

⬐ 2Gkashmiri

yeah, post 2019, after the "internet" ban of 9 months was lifted, i was given the privilege of accessing 2G internet so that is the "backstory" of this username....

⬐ mfrw

Its relatable; I belong to the same part of the world.

⬐ 2Gkashmiri

btw, i was discussing tldr with a friend recently and to find you a maintainer, nice work dude.

⬐ 2Gkashmiri

oooo.. nice... don't know about you but i've been online for too long and been burned many times to not maintain a single identity online.

like, back in 2010 i think, this guy warns me "WTF are you tweeting incendiary stuff with your name..." and i was like "holyyyyyy sh...."

since then i have maintained watertight identities across any public platform i join with handles i never recycle".

i obviously don't post incendiary stuff nowadays because of being busy with work but the habit has stuck.

⬐ schmorptron

The Intel iGPUs have open source drivers in the Linux kernel, don't they? In that regard they're even better than amd since you don't need to get proprietary drivers to do compute stuff.

⬐ wmf

This Intel-approved not-a-review isn't a good look for LTT.

⬐ cptcobalt

Strong disagree. The videos Intel did with LTT and GamersNexus reveal inner workings of some of the GPU architecture. Certainly a marketing-focused presentation targeted at tech-savvy folks, but some of that narrative is required to really set the stage for a new entrant into the field, since there's no preexisting story, and existing mental models of Nvidia & AMD GPUs don't mesh so cleanly into the Intel GPUs. Knowledge will advance much faster with some marketing then with pure lab tests. And it's obvious LTT and GamersNexus plan to test these to the ends of the earth, like they do AMD & Nvidia cards.

⬐ gjsman-1000

LTT is trying to build a lab and grow - but to be honest, if I was a qualified engineer, why would I go to work at the LTT Lab? All I would get to do is review other products and then have my findings boiled down into 15-minute summaries where nuance isn’t a strength and you might get lucky and be on a clickbait video thumbnail. Boring compared to actually building products.

If I’m an engineer from, say, Intel, I’m not moving from Oregon or Arizona to Vancouver for it.

⬐ kadoban

The output of lab isn't going to be just videos, and they don't appear to be having any issues hiring.

⬐ sandworm101

Unless they break into the lucrative consulting/reverse-engineering game. Rather than videos, they might start writing reports paid for by actual manufacturers, reports that would influence realworld products

⬐ theshrike79

Why would the only output be a 15 minute video?

I'm expecting they'll do Anandtech/iFixit level writeups along with the video.

⬐ bobdvb

Yup, he's already said they're planning to publish extended write-ups.

I imagine it'll be something along the lines of ServeTheHome but with wider topics.

⬐ favorited

You'd also be going from working with fellow engineers to working with influencers. Look at how often they do a "we lost our archived data" video – their whole schtick is making videos featuring how many hard drives Seagate sends them rather than how to make informed decisions.

⬐ Dylan16807

> Look at how often they do a "we lost our archived data" video

Twice in a decade?

The first time they lost no data, the second time they lost a small amount because they weren't monitoring failed drives.

Considering it's data they don't care about enough to back up, that doesn't seem so bad.

⬐ AlotOfReading

Well they don't need to hire thousands of such people, they need like a dozen. I'm certain they can find that many competent engineers in greater Vancouver, even if some of their peers have left to the US.

⬐ CoastalCoder

Yeah, I wish LTT was more obviously independent with these Intel videos. LTT's credibility is it's main value IMHO.

(Just to be clear, I'm not saying that the content itself is highly sus. It's more that LTT's heavy airtime with lately with that Intel dude seems a bit odd.)

⬐ denkmoon

How so?

⬐ wmf

In general when a company is controlling what you can say, they're doing it to make their product look better. Publications that aspire to actual journalism shouldn't participate in that.

⬐ jamesgeck0

LTT is generally pretty transparent about any restrictions companies want to put on previews/reviews. Intel’s conditions are spelled out multiple times in the video, and LTT aired a longer discussion with these folks on their podcast that contained a fair amount of Intel criticism.

This is 100% Intel trying to set the correct level of expectations for what will be a extremely competitively priced GPU with major compatibility issues at launch.

⬐ jeroenhd

They're transparent enough to be trustworthy for reviews, but not they're not really transparent about the influence a company might have. They will do videos for a manufacturer reviewing a product with a given set of rules, but don't always disclose their rules. They argue that their viewers know that if "showcase" is in the title they're being paid for the product (though they often omit "showcase" from their clickbait titles before adding them back later) but their custom definitions for content types are hardly an industry standard, let alone well known to the viewers.

They do disclose what companies send them what, at least. They're better than most YouTube reviews in that sense.

Their current content is always entertainment first, sales second, reliable reviews third. They're a media company, it's the way they do business.

I have no idea what they were trying to do in this video other than show off some cherry picked samples. I'd much rather have seen a quick table of stats next to the unscripted interview during their podcast to be honest, but that doesn't bring in any cash.

⬐ sirtoffski

Overall I do agree with you, it does make reporting seem very “sponsored“ and disingenuous.

However, an argument could me made that “We have the first look at the new technology. We’re obviously very limited at what we can legally say or show, but nevertheless here’s a very limited sneak peak” is still genuine journalism. They’ve reported as much as they could and did not hide obvious Intel participation, etc.

The video was very underwhelming and cringey imho. But to informed audience it probably wasn’t misleading or false - that’s how I felt after watching it.

⬐ mardifoufs

Id agree but LTT has been very critical of intel and at some point intel was literally the butt of the joke on the channel for how abysmal their performance and thermals were. Even the previous video right before this one on the Arc GPUs was basically a huge call out, which was obviously taken seriously by intel as they have explicitly addressed/fixed the issues that were raised in that video. This is basically the best case scenario you can get as a customer, as far as commercial relationships go. It's so much more constructive for everyone involved to allow intel to also "respond" to the criticism by showing how they improved things.

( It also goes to show the type of influence LTT can have when they bother to send actual engineers to showcase and prove those improvements. That's something they probably wouldnt have bothered with if they knew LTT was just going to parrot the corporate line if they showered him with enough money.)

⬐ arein3

Don't think he cares. Look at the yt thumbnails.

⬐ jeroenhd

This is a topic many youtubers discuss. The stupid thumbnail faces drive views and bring in sales. Everybody knows how dumb they are, but it works. Same with the clickbait titles: unless they're outright lies, they help immensely in getting a video recommended by the YouTube algorithm.

There are a whole range of algorithm tricks that are necessary to build a business on YouTube. For example, the first 30 seconds or less are crucial to grab people's attention, so overloading them with entertainment or promises about what's coming is essential for maintaining viewership even if the quality of the video overal suffers.

I haven't yet caught them doing the more egregious algorithm tuning (intentionally putting mistakes and spelling errors in graphs to drive comments, for example, or spamming polls to YouTube to get their engagement stats up). The YouTuber face is kind of a turn off, but when I watch a video I don't watch it for the thumbnail anyway.

⬐ arein3

I don't watch videos from such guys.

The over enthusiaatic reactions feel cringe and fake.

The thumbnails are dumb at best.

It's a person that is in this for money or has no self respect that would do such things, and I don't value their opinion.

There is better content, it's not needed to put up with this disrespect for the viewer.

⬐ smcl

Yeah the thumbnail gurning is silly, but by now it's a lost battle - they're an integral part of many popular channels. At least it can be a source of amusement, I frequently think of this tweet by Rob Whisman: https://twitter.com/robwhisman/status/735281634656669696

⬐ silisili

Basically all tech reviewers, LTT included, are garbage because of this arrangement.

To have a day one review, you have to get a prerelease device. To get said device, you have to stick to mfg talking points. It's an awful circle of scummy companies and scummy people.

Really wish there was a reviewer who just buys shit with their own money AFTER release, and tells it like it is. Some may exist, but none of the household names have the integrity to do so.

⬐ bobdvb

The problem is that, as Hardware Unboxed and others have demonstrated in their disputes with nVidia, if you're not releasing reviews on the manufacturer's schedule (e.g. samples available early and then following embargos for releases) then there's no viewership for your videos.

Literally if you haven't made and released a video in the two days after the embargo drops, you're not going to get substantial views, at least not enough to be a household name.

So, even if you are independent, you play the game and you try not to p*ss the vendor off while also maintaining independence.

Ultimately, the Intel videos by LTT and GN weren't a full review, and both were previews. But importantly, they were previews in a way that we really aren't used to big corporates allowing, and that's a positive change. While both channels were clearly being friendly, I don't think there was anything shill about it. The vendor set the bounds of what they could show in exchange for the right to share what was allowed, but I didn't see them reading from the hymn sheet. Heck, Intel spoke openly things that previously nVidia and AMD would have been trying to hide or avoid, it's refreshing.

You take the videos for what they were, a preview under specific conditions. Later on there will be full reviews, unconstrained benchmarks and time for a full critique. But I bet with great certainty, most people won't ignore these preview videos, or any "day zero" reviews, in favour of channels that don't work with Intel PR.

⬐ rawoke083600

>Basically all tech reviewers, LTT included, are garbage because of this arrangement.

YMMV. But I won't say it's garbage. I do get value out of this and similar "reviews"

⬐ theshrike79

LTT is actually building a specialized shop just for testing hardware, with actual experts running it.

⬐ 0xy

Don't forget said pre-release devices are often different, sometimes substantially, from the released version.

LTT and many other reviewers don't even go into depth and often get details critically wrong, such as falling victim to Nvidia's GTX 970 "4gb memory" scam, because they didn't know how to test VRAM and discover 0.5gb of that memory was severely degraded in performance.

They also don't appear to have any process to vet people they take money from, given they're partners with Gigabyte who dumped PSUs onto the market they knew were dangerous, and several exploded as a result.

They aren't reviewers because they have a conflict of interest. If they fairly review products, companies will not send them review units.

⬐ asiekierka

The only world in which fair product reviews lead to no product review units whatsoever is one in which no company ever actually considers their own product to be of quality.

Ultimately, follow-ups are rare because they don't sell - the benchmarks which get referenced are the ones which spread the widest, which are the ones closest to release, which is when usually the most people are interested in a given product.

⬐ mkaic

You're describing Gamer's Nexus pretty well I think -- easily the highest quality and most transparent/trustable PC hardware reporting I've personally come across. They frequently burn bridges with manufacturers because they refuse to stay quiet about major flaws in products.

⬐ MikusR

Gamer's Nexus also did multiple videos with the Intel people. And LTT also burn bridges.

⬐ mardifoufs

I kind of disagree. I like the gamer nexus reviews, but the constant stream of perpetually angry/ranting videos also gets tiring at some point. Like yeah, controversies and angry call outs are fun but what's the point?

I prefer the ltt approach here. They went pretty hard on Intel on their original Arc video a few weeks ago, but the fact that they are known for not burning bridges made it worthwhile for Intel to respond and showcase the improvements they have made since. As I said in another comment, they wouldn't have bothered if linus could've been just paid to update his review to make intel look better. But I also don't think Linus would have had that type of influence (enough for a corporation like Intel to engage with his complaints directly ) if he just kept burning bridges instead of keeping channels open.

(Gamer nexus is far far from the worst in that regard, and they too had a visit from intel engineers. But I still think the LTT approach is just different, but I dont think that means they are just corporate shills.)

⬐ rawoke083600

Ja he is not bad (TechJesus), but he can "ramble" a bit.

⬐ mardifoufs

Absolutely, the entire GN crew are very good at what they do. I just think that they aren't necessarily better than LTT just because they appear to take pc gaming component business very seriously. When in reality, the methodology itself is nothing groundbreaking or really special, they just make less jokes than LTT

⬐ Y-bar

> the constant stream of perpetually angry/ranting videos also gets tiring at some point

Hear me out here. Maybe it is warranted and we ought, instead of directing the criticism towards the messenger or those who expose problems, direct this towards the manufacturers which produce subpar products?

⬐ mardifoufs

I'm not saying it's not warranted, it's just a different mindset I guess. I don't usually consume tech content to get mad or to find drama. I don't see the poing of yet another video about how bad alienware/oem builds (that no one in GN's audience would buy anyways) really are. It just feels like preaching to the choir over and over again, especially when it's often a bit rambly.

I'll still totally watch GN reviews of specific products that I might purchase. But that's kind of my point, LTT and GN fill a different niche. If I'm not shopping around, at the very least LTT videos are more "entertaining". And even then, while their reviews may be less serious, they usually still end up with very similar conclusion than GN regardless of who is sponsoring them.

Linus can be cringy but imo that's better than the current trend in some PC tech channels to take completely trivial stuff, like say the 3090ti being not much better than the 3090 (oh no!), super seriously. It's useful info don't get me wrong, but it doesn't warrant an angry reaction or outrage.

(If anything GN is really not that bad in that regard compared to other channels, but it still happens sometimes.)

⬐ bobdvb

Yea, I have to agree, I look at GN videos that pop up in my timeline and have to decide if I am in the mood for that kind of negativity. It's like trying to decide if you're going to put the evening news on and face all the pain and suffering that's going on in the world.

⬐ Dylan16807

What specifically would you change? They talked about the important failings of the GPU drivers, and it's not worth giving many specific numbers when they're likely to change before release. They also can't evaluate the whole package properly until a price is announced.

There's not really anything intel could be hiding here that's worse than the big performance issues they acknowledge. And I'd much rather have this kind of preview than nothing.

⬐ wmf

I think I'd prefer nothing, but that's just me.

⬐ w-ll

Just Walk Away From The Screen

⬐ Teever

So just to be clear, you don't like the content and wish it didn't exist, but can't really articulate much beyond that?

This is some 'old man yells at clouds' level posting man. Why do it?

⬐ LoveMortuus

tl;dr: Received LTT hoodie as a gift (the gift giver paid almost 2x the price due to shipping and import), the quality of materials is so bad that since trying it on, I haven't worn it since.

I used to like their videos because they seemed honest and somewhat transparent. But about a year ago I received a LTT branded hoodie (WAN hoodie) as a Christmas gift and I was seriously shocked but the quality. It's July now which means that is been more then half a year and I haven't worn it once since trying it.

The inside of the hoodie is extremely rough so much so that it could be unpleasant to the touch. The zipper is flipped, which means that you have to use the other hand to zip it, which here where the position of the zipper is basically standard is perceived as just basically wrong, at first I thought that maybe it was a manufacturing defect, but there seem to be many people that report a flipped zipper and when I was trying mine on quite a lot of lose threads were just falling out.

This was a real shock to me, especially because LTT always talks about how quality is important and how they won't release a product if it isn't good.

Sadly I can't return it, because it was a gift and I've been told that the import costs and shipping for the hoodie were almost the cost of the hoodie.

⬐ rawoke083600

To be fair, I think they are changing that or the focus on quality. Looking at their recent videos it looks like they constantly on the hunt for better manufacturers and even do their own testing like color-bleads washing.

⬐ LoveMortuus

You're right. Their adventures regarding the development of their screwdriver does give you hope that it might be good, like the tools used to be, where you bought them and they stayed with you for the rest of your life.

⬐ IntelMiner

Have you considered just...contacting their support team? It's Linus Media Group, not Versace. They pride themselves on good customer service

⬐ LoveMortuus

I have considered it. But there are three problems that I encountered:

1. It was a gift, so it just feels wrong, which is of course completely subjective and on me.

2. It being a gift I don't have the receipt for the hoodie, which could cause issues.

3. Shipping again. For those wondering, I'm from Slovenia, Europe, so it is quite away from Canada.

⬐ proto-n

A quick note about zipper sides: customs are quite varied on this in different countries. From shipping cost I assume you live far away from canada, maybe it's just that your country has it the other way around?

You can't really fault ltt for that.

⬐ LoveMortuus

You're correct, I'm from Slovenia, Europe, which is basically on the other side of the planet. Regarding the zipper, you're correct, it's not a thing that people think would be different around the world and thus don't pay much attention to it when buying stuff on the internet.

⬐ AlphaSite

They also did another good unscripted section on their podcast, which I enjoyed, Ryan sprout and friend are both fun to listen too and seem to happily share details.

⬐ blackwatcher

hi how are you

⬐ blackwatcher

HandOfBlood

⬐ blackwatcher

Kelvin und marvin

⬐ zathan

- Similar performance to AMD and Nvidia in optimized games (tier 1 games)

- Similar performance to AMD and Nvidia in Vulkan and DX12 games (tier 2 games)

- Inferior performance to AMD and Nvidia in DX11 and below games (tier 3 games)

- Positioning and pricing will be based on tier 3 games

Intel GPUs can be a real bang for the buck for Linux gamers due to DXVK being used with most games. It will also be interesting to see if DXVK can also improve Intel's performance on Windows games.

⬐ ceeplusplus

Intel's not the only one who has had inferior performance in DX11. AMD's DX11 driver suffers in drawcall heavy scenarios, particularly in dense open world games like Witcher 3 [1]. In the old days they used to underperform even more. It surprisingly is quite hard to develop a performant DX11 driver.

[1] https://www-purepc-pl.translate.goog/test-ryzen-7-5800x-i-ra...

⬐ zathan

Very interesting link. I forgot about it since I mostly use Linux and DXVK, but on Windows and DX11 tittles it was (maybe still is) very common for AMD GPUs to need a lot more horsepower than their Nvidia counterpart to reach similar performance levels.

⬐ ThatPlayer

Zink (OpenGL on Vulkan) could be interesting, since OpenGL isn't brought up at all, but probably falls under "DX11 and below".

⬐ kllrnohj

> Similar performance to AMD and Nvidia in optimized games (tier 1 games)

Similar performance to AMD & Nvidia's lower midrange cards fwiw. Intel's top tier sku isn't aiming very high, only seems to be targeting something like the RTX 3060 (non-ti).

Fine enough place to start, but not exactly the most exciting thing even if the price turns out to be great.

⬐ LordNibbler

Long-term this great news for the consumer that there is a third competitor in the discrete GPU space. If the tier 1 games have competitive/same performance then I'm optimistic that the HW is solid and the gap needs to be closed via updates on driver software.