Intel Xe Graphics Preview v2.0: What we know about Intel's upcoming GPU

[Julio Franco]

Intel is developing discrete GPUs for gamers, professionals, and servers, and they're all slated for release this year or coming in 2021. Intel's cards will either be the long-awaited saviors of a stagnant market, or they'll underperform and flop miserably (no pressure, Intel PR person reading this). This is our second round of investigation into Xe.

Read the full article here.

https://www.techspot.com/article/1975-intel-xe-preview-v2/

 

[benderOS]

Why do I have such a felling, that these gpu’s are going to be either worst price/performance offer in history, or something totally unimpressive like 1650? - and there’s no other option with that...

 

[noel24]

LOL, that RAMBO cache is expendable...

 

[Vulcanproject]

Software is one of my main concerns. AMD and Nvidia have large driver teams, that can rely on decades of previous experience with baked in fixes and tweaks for thousands of legacy games included in every release. They also enjoy wide community support for obscure game fixes, especially Nvidia.

Intel have to assemble a huge software team and roll out rapid driver improvements to build a good reputation with any potential customers for these graphics cards. If the hardware is good but the drivers are immature and broken, all the effort to move into this market will be for nought.

 

[QuantumPhysics]

Intel already has a good thing going since they make up the majority of the market's integrated graphics which people are using without a GPU card in most situations.

All they really need to do is target the low end, medium end and high end with cards that outperform or match the 1660 Ti, 2060 Super, 2080 Super and the 2080Ti.

Ultimately they have to beat the competition at a slightly lower price. The rest will take care of itself.

 

[psycros]

Intel already has a good thing going since they make up the majority of the market's integrated graphics which people are using without a GPU card in most situations.

All they really need to do is target the low end, medium end and high end with cards that outperform or match the 1660 Ti, 2060 Super, 2080 Super and the 2080Ti.

Ultimately they have to beat the competition at a slightly lower price. The rest will take care of itself.

See the post about drivers above - that's a huge potential stumbling block. Also, the only hope for reasonable price points on this new silicon will come from AMD's half of the equation.

 

[Azshadi]

Intel already has a good thing going since they make up the majority of the market's integrated graphics which people are using without a GPU card in most situations.

All they really need to do is target the low end, medium end and high end with cards that outperform or match the 1660 Ti, 2060 Super, 2080 Super and the 2080Ti.

Ultimately they have to beat the competition at a slightly lower price. The rest will take care of itself.

Not even that. The whole idea of beating the competition at la lower price blew up in AMD's face with every GPU they put out, targeting the low end exclusively is another rabbit hole that they would need to avoid, it didn't pan out for AMD I really have my doubts about a new player taking over that bit from others that have been in the business for way more time than them. Their ace in the hole is still the IGPU that power pretty much 9 out of 10 businesses (utrabooks) the low power parts cater to and I'm pretty sure that's what they would aim for since that segment is still "unchallenged" by AMD or nVidia and by that I mean that they move more inventory out for that segment than both AMD and Nvidia. What would they need to do is correctly fix pricing based on performance. I for one wouldn't mind paying 30% extra for 30% more power (I know it's far fetched seeing that Intel has shareholders in mind first). As for the software side I'm not bothered, out of all 3 players I'd expect them to top nvidia's support seeing that Intel is more capable of spending money in that front than AMD+Nvidia. All that's left would be the reception of the public for the products, IF and it's a big IF they price them accordingly we could see a stable GPU market for once (that's what I'm rooting for at this point). The only way Intel can botch this up is by their own hand anyway.

 

[amghwk]

Until we have a hard head-to-head reviews of this card against the Nvidia and AMD giants, we can pass this long article without reading.

 

[PEnnn]

Ultimately they have to beat the competition at a slightly lower price. The rest will take care of itself.

Their track records when it comes CPU pricing is not a good one...why would their GPU p;ricing be any better??

 

[ShagnWagn]

I support their efforts, hoping they will bring more standardization to the industry. Intel makes their chipsets so many different manufacturers can integrate their products in the market. I hope the same is for monitors and (legacy) game support. They are going to need to beat nvidia and AMD at all points to penetrate the market. There is a steep mountain to climb, but they have the funds to make it happen. It will cost them much more given they don't have the experience.

 

[SG736F6]

This is great! I welcome more competition in the GPU market. If Xe can compete, prices will likely drop across the board.

 

[Rock Dirty]

This is going to go down in flames bright enough to light up the night sky.

I will bath in the majesty of its glow.

 

[Boshum]

Intel already has a good thing going since they make up the majority of the market's integrated graphics which people are using without a GPU card in most situations.

All they really need to do is target the low end, medium end and high end with cards that outperform or match the 1660 Ti, 2060 Super, 2080 Super and the 2080Ti.

Ultimately they have to beat the competition at a slightly lower price. The rest will take care of itself.

Yeah. That's all they have to do. No sweat. Should be a breeze.?

 

[Lionvibez]

Their track records when it comes CPU pricing is not a good one...why would their GPU p;ricing be any better??

They are the dominant 800 pound in gorilla in the CPU market so they can charge high prices.

They will be coming in as the underdog in the GPU market and won't be able to charge top dollar with no previous history of dedicated GPU products. There position won't allow them too much space when it comes to pricing.

 

[QuantumPhysics]

Yeah. That's all they have to do. No sweat. Should be a breeze.?

Yup

 

[Irata]

Yeah. That's all they have to do. No sweat. Should be a breeze.?

Realistically, they need to do two things: make good enough low and mid range GPU and convince OEM to use their product rather than the competition's.

I think that they are good at the second part has already been established.

Do they need a 2080 / 3080 beater? No, not really. Low and midrange OEM sales is where the volume is and they could hurt nVidia the most. Dedicated server / data center GPU would be a second attack vector.

As depriving their competitor of income is another tried and true (and successful) Intel strategy, my bet is that this is what they will go for first.

 

[meric]

AMD has still not delivered what many people ask from them: capable APUs. I'm talking about "mainstream" level of performance not the lowest end hardware. Their best APU gave lowest end GT 1030 level of GPU performance at best. Many people still wait for an APU with mainstream level of performance. Is it so hard to design an APU with Ryzen 3500/3600 + Radeon RX 580/5500XT or even a 5600? If intel offers products like this, they will be successful.

 

[Anatun]

I have a very good feelings on Intel's Xe GPU efforts due to the leaked benchmarks. The Rocket Lake Xe IGPU is indeed impressive. It seems to be using 1/4 th the power of a similar performing Nvidia GPU. The power budget is the main reason why Gaming laptops underperform. If Xe is that power efficient like the leaks are showing, it will revolutionize the mobile gaming. Laptops won't need to be thick and heavy for game. Looking forward to it...

 

[CharmsD]

Do they need a 2080 / 3080 beater? No, not really. Low and midrange OEM sales is where the volume is and they could hurt nVidia the most. Dedicated server / data center GPU would be a second attack vector.

If Intel doesn't up their high-end they may as well stay on the porch. They're in low end (portables) and with Atom for other devices. They're deep in compute (Q3-'18 here) and big datacenters. Building these new stacked GPU's with 9 kajillion EU's, sort of, kinda, points out an intention to do that. The bottom end [*should be easy] if they can increase performance to or above 2080's, it *should be comparatively easy to build an add-in card, 'down'. Laptops present power and thermal challenges that are increasingly difficult, if Intel does ramp up a combo IGPU + GPU that could put their mobile solutions over the top for some time.

Having the faster PC, GPU would give them all important bragging rights, marketing geniuses are going to boast anyway, but they really need to earn it..

Intel's finesse in evaluating yield/wafers for long-term profitability could entice them to ditch 'Xe' for something different, such is the Intel way. Nvidia is absolutely not sitting in wait, I don't think we've seen the last of TU- chips, EVGA's speedy-cool KO stays SOLD OUT.

 

[Irata]

Look at how they conquered the server market with Xeon - they were the cheap good enough alternative to Sun & the like - at first.

They may go for the high end later on, but if they manage to push nVidia out of both OEM laptops / desktop systems (most if those do not even have 2060 class GPU) and servers nVidia will lose a lot of revenue which will greatly impact their R&D.

It would actually be good for Intel if AMD can compete on the high end with big Navi as that way, OEM would not need to fear any reprisals from nVidia shortages for high end nVidia cards.

 

[pp luh pee yoo]

Nice they got their Execution Units in order...
Hate to tell you, so do all the other guys.

I see it as a basic nuthin' article. You aint got a card,
you aint put it to the test.

What about a firm afirmatittie on the card's appearance for the
droolie Blue Boy masses ??? HUH? Would that be real news?

 

[Evernessince]

See the post about drivers above - that's a huge potential stumbling block. Also, the only hope for reasonable price points on this new silicon will come from AMD's half of the equation.

Maybe Intel's large iGPU install base helps them with compatibility?

 

[axiomatic13]

(30 years exp in the server industry) I have ZERO expectations on Intel's ability to execute. The Xeon Phi cards still have incorrect instructions on their "How to install Phi" support page.

 

[CharmsD]

Look at how they conquered the server market with Xeon - they were the cheap good enough alternative to Sun & the like - at first.

That's almost like saying "Intel knows what they're doing".

They may go for the high end later on, but if they manage to push nVidia out of both OEM laptops / desktop systems (most if those do not even have 2060 class GPU) and servers nVidia will lose a lot of revenue which will greatly impact their R&D.

IMO, now is LATER. Nvidia isn't waiting for Intel's June, July, or an "opps we have a problem" excuse, they're rolling ahead with a nex-gen slap-your-face, super-duper RTX/DLSS silicon intended to be enough for a five year run. The current crack in time is when Intel has an opportunity to leverage their wares, the next-best available time slot is pre-christmas.

And they need software, tons and tons of software, fortunately they are very good at software, at times. I have confidence about 50.00001% Intel won't ef*** [starred myself] this up.

It would actually be good for Intel if AMD can compete on the high end with big Navi as that way, OEM would not need to fear any reprisals from nVidia shortages for high end nVidia cards.

This is a different kettle of fish, I don't know anything about relationships between NV and OEMs.

 

[amstech]

Looking forward to Steve's review.