![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
This is a good thing. Good ole Pat is on record saying that computer graphics has no impact on the future of computing. You need look no further than him to see why they're not even attempting to compete in the GPU space and if their mediocre 160 & 192 bit graphics cards aren't a success they will relegate the future iterations of their GPUs to on CPU die only.
Never understood why people post third-parties regurgitating press releases when you can just go to the press release.
![Zarathustra[H]](/data/avatars/m/9/9298.jpg?1723407683){kind=avatar}
Pat failed at all things engineering, at all points in his career in which he was the lead. He is the architect of Intels failures. He was part of a team early sure. His time as CTO set intel up for a decade of failure which he skipped out on. Then he returned and every single thing they have released with him at the helm has massively under performed. OR was defective.
Gelsinger was a hellaciously good CEO at VMware - I worked there under him, and with his direct reports, and he was one of the best they'd ever seen and - honestly - one of the best I'd ever worked with/for. He's the reason they grew to the point that Dell/Broadcom/et all wanted them that badly. He knew how to cut fluff (why VMware owned Zimbra, or SlideRocket...) and concentrate on the core parts of the company, and he grew them ~massively~ - and then concentrated and pushed for quality when that was a major issue.
He wasn't far off the 5 year mark. He turned a 70 dollar stock into a 20 dollar stock. When you divide a companies stock value by 3. You tend to get fired. At least the US gov allowed him to retire.
By all indication all he has really been doing is canning projects. Such as Jim Kehlers royal core stuff. He also got ride of Raja on the GPU side... no matter what the prevailing thoughts were on him. For the most part all he has done since he came in was get rid of big names in CPU GPU and Fab. Ultra is 100% on him. He is probably going to take the blame for whatever follows it up as well.
He was massively underestimating the future then - but lets be honest, lots of folks underestimated what GPUs would become too (lots of places considered buying Nvidia early on and didn't). Vector math being useful for things other than graphics and HPC workloads wasn't entirely expected; I worked in the HPC space in 2002-2005 and we thought massive low-power parallelism was likely a future path there more than GPUs (see: Blue Gene /L)
Oh he failed. I was a strong defender of him, for certain - and I thought (think? Hope?) they're on the right path in the long run here, it takes time to build a new CPU design and get there. Jim Keller went back to AMD in 2012 - Zen released in 2017. 5 years for a new design to tape out and get to release. The arrow lake successor is what I figured we'd see as the first real "Gelsinger led" solution, although I'd hoped Arrow Lake would be better too and show improvement on the way. I got worried when Keller's Royal Core was canned, and ARL doesn't show any notable improvement yet... I still figured he had one more release to see if he'd been right on that choice. Now I fear they're fucked, and can only hope that what they've been working on pans out. I don't want a single vendor market with just AMD any more than I want a single vendor market with Intel.
I honestly figured he had till the ARL successor came. If that was strong - groovy. If not - well, he had a full product dev cycle at that point, and if he couldn't get a good one out, then he needed to be gone. 5 years from start of design to release isn't abnormal for a CPU.
FWIW, all of those were designed before he came back, Alder Lake was through tape out in 2021 when he arrived (and ES/QS had been out for a bit already), Raptor lake was taping out at that point, and only 14th gen and Arrow Lake were in design still.
Pat left in 2009 and came back in 2021 - I'm confused a little on your timeline here for 11nm+++? Did you mean 14nm (the half dozen Skylake variations)?
Started in February 2021. Which products was he supposed to influence? 11th gen was out, 12th gen was in ES/QS sample (taped out fully and in pre-production), 13th gen was in tape out, 14th in design, and 15th (Ultra) was in early design with Royal Core. Now we can argue all day long that he should have let Keller cook (when has betting against Keller ever worked for ~anyone~?) and that his E-core only dream is nuts (I think there's a use for both in DC and home uses), but arrow lake is the only one he could have had any influence on really, and not enough to truly define what it would be. ARL successor is his baby 100% on the standard 4-5 year timeline from start to release; we don't know what that will look like yet (but I'm worried).
Royal Core has me worried. Raja - eh, we'll see - he's better than what they had before for sure, and I'm not sold on their GPU business...
Well lets see what people say now that he is gone. Pat is about to get blamed for the shit show the next CEO inherits.
Um. Gelsinger left in 2009. That was the release date of the I7 960, which was on 45nm. Sandy Bridge was 32nm. Haswell was 22nm. Skylake was 14nm. EUV lithography went truly mainstream in 2018 (https://archive.ph/Fd09z - no paywall on the MIT review article), which is what got us to below 14nm, as the original processes used wouldn't scale lower.
He was previously responsible for Itanium, this is all one should need to know. But Intel was still profitable with Itanium because others (HP) paid for it. This was revealed during HP v. Oracle lawsuit.
I disagree with the strategic part. He achieved that long-term company survival is now guaranteed by US taxpayers.
Arrow Lake certainly didn't help but is not that relevant either. Most of Intel revenue is in servers and datacenter (same with AMD), the gaming market is rather small in comparison.
5 years from initial design to release. When Pat started in 2021, 11th gen was out, 12th was taped and in QS/ES, 13th was in tape-out, 14th was late design, and 15th (Ultra) was in some stage of mid design, with the successor in early design. He had influence on 15th, and 16th was the first one he'd have real control over. For better or worse.
To be clear I'm not saying some of the things he cancelled didn't make sense to cancel. I just wonder how his canceling, pushing up, splitting CPU teams into those working for non Intel fabs and those still working toward Intel fabrication. One rumor that seems pretty well supported by recent products is that their is two groups of Intel CPU engineers, one group hates E cores another that supports them... this resulting in odd decisions like SMT being disabled. Also AVX512 being dropped at one point.