Absolutely correct. Core was the next desktop architecture after P4, which is why I specified it as next in line after P4 chronologically, not architecturally as you correctly pointed out.
Just a nitpick but Pentium M is more derived from Pentium III than 4.
I doubt that Raptor Lake Refresh (if any) would be on Intel 4. If it were, be prepared for disappointment since there might be clockspeed regression.
Problem... or expected difficulty. Scaling down past where they currently are is getting massively more difficult. Witness other leading foundaries having their own problems with achieving better total processes and many others just throwing in the towel along the way. As it stands, all of the foundaries on the leading edge are essentially peddling half and quarter steps between node variants to eek out even the tiniest of improvements to fill the long gaps between full nodes. I don't see Intel as being in trouble as much as I see the crowd catching up to them and occassionally gaining a small lead. We may bag on Intel's 10nm process, but it was often within spitting distance or even better than competing nodes for absolute circuit density. When fed reasonable power, it returned reasonable performance that improved with each revision.
If you think Intel is doing so much worse, look at their competition. Samsung has struggled to stay within a full node of TSMC and Intel, underperforming all along the way for years. TSMC is experiencing delays and having to issue less aggressive subnodes on N3 (N3e being one) to keep customers happy. Who else is anywhere near these three?
Intel's schedule is, as usual, overly optimistic. They aren't falling behind on process tech though.
If they increase the internal caches (ALL OF THEM) and the various instruction buffers too, does that count as a refresh or a new architecture?I doubt that Raptor Lake Refresh (if any) would be on Intel 4. If it were, be prepared for disappointment since there might be clockspeed regression.
Okay. This makes more sense. I think this Intel delay rumor was spurred by the rumor spread around last and this week that Apple is purchasing all of N3 for the next 2 years. I don't remember if they're still on N3, but I was wondering why this would be so if others were going to N3E. I've read some conflicting information on MTL-S but I think we all know it's not coming to desktop at this point and that may have been the plan all along. I get Intel wanting to push Arrowlake S on mobile only in 2025. They've still got mobile on lock but are facing stiff competition from AMD's new hardware. AMD however cannot provide as much raw product to manufacturing partners as Intel can. They can take a potential small beating or market loss there only to come roaring back with Arrowlake S. Though mobile MTL-S should provide them enough breathing room. Gelsinger stated MTL was a game changer. Whether this was a white faced lie or he really meant it is to be seen.
Refer to Nostra for any future releases.Reminds me of one time when someone came up with such an article to support a claim made on Anandtech forums... only for us to realize the article was based on the Anandtech forum claim
Who said that? I was drinking last night but I wasn't that far gone. What you say here is what some of us originally thought. The GPU makes more sense to be fabbed at TSMC because they excel at low power, relatively speaking. Intel's commitment was solidified earlier this week with the announcement of their next gen GPUs. Whether Koduri is still involved in the project is another matter and question.Again I doubt Arrow Lake cores, whatever form they take, will be on a non-Intel node. The GPU? yes. CPU cores will be on an Intel process, GPU on TSMC, and everything else? Depends on who is cheaper.
Moving to TSMC for a client core would be a disaster for Intel in many ways. So I firmly believe they won’t do it. Further, they have no reason to.
Just putting this here to stake my claim. We shall see how it plays out.
If they increase the internal caches (ALL OF THEM) and the various instruction buffers too, does that count as a refresh or a new architecture?
Still unexpected for Intel it seems. To me it has become completely ridiculous just how unprepared Intel repeatedly appears to be to every single delay that predictably comes. For all that Intel still fares surprisingly well in a way, but I'm not sure anymore what that whole (constantly overpromising and underdelivering, never on time) show's purpose is.Problem... or expected difficulty.
These products are more race steam locomotives with their boilers close to explosion than diamonds.But AMD and Intel have pressured each other lately and both have responded to the pressure by making diamonds.
If you're talking about L1/L2/L3 then I think that would be a refresh. If you get into micro cache/reorder/branch order buffers, registers, loads/stores, and similar then I'd call it a "Tick."
New architecture generally involves increasing decoders and/or ports, which in turn requires the above structures to be increased/optimized to adequately feed the data.
Sometimes Intel would get things a little out of balance front vs. back end and fix it in later. For example, Haswell went from 6 to 8 execution ports but kept decoders the same. I'm sure Intel's models showed good IPC increases by opening up the back end. Then the front end became more of a bottleneck and they added a simple decoder with Skylake.
Then a return to the back end by going 8 to 10 ports with Sunny Cove. And finally with Golden Cove another simple decoder and 2 more execution ports were added. The buffers, registers, OoO order logic seems to be revised as the front/back end is opened up to support the greater data flow.
It *seems* like Intel rectified the bandwidth issues with Golden Cove by increasing L2/L3 with Raptor so Raptor Cove may be "balanced" and won't benefit from additional tweaking. But their models may have identified a few simple changes to allow for a few more percent IPC as well as some frequency increase.
But if we look back at all of the refreshes of Skylake because that's essentially what they were, all Intel did was add cores and increase frequency. Doubtful they will add cores (to the top of the stack) but I'm hoping they may be able to bring down power across the stack.
For example, while the 13900K can run 5.5GHz all cores, outside of a car radiator sized custom loop it just isn't sustainable with most of the silicon they are selling at 13900K from what I've been reading and experienced myself.
So let's assume the Raptor refresh can run 5.5GHz at the VID the "old" Raptor could run 5.1 or 5.2GHz. That would instantly be an upgrade for anyone moving from 13900K Raptor to Raptor refresh that doesn't have a monster cooling solution even though the label on the box would show the same specs!
Problem is how does Intel sell this? Right now for those of us in the "know" the 13900KS is more valuable for it's ability to run same clocks as 13900K but at lower voltages. No guarantee of course but you basically pay for a higher binned part. Intel "sells" this by slapping "6GHz" on the box. That number is meaningless outside of benches because other cores always drop that top number.
So in conclusion Intel is in a bit of a bind with this refresh. On the lower end parts below the 13900K/KS they can add some frequency across the board. But at the top of the stack there is really no where to go because they have that segment so finely binned with the 13900K and KS.
That's why I find this refresh interesting. I'm curious as to how Intel will proceed. And I hope Zen 5 is right around the corner and it is all we're hoping it is. Love seeing Intel's feet put to the fire and see how they respond. But AMD and Intel have pressured each other lately and both have responded to the pressure by making diamonds.
These products are more race steam locomotives with their boilers close to explosion than diamonds.
The silicone pieces are fried and hundreds of watts of heat pumped through them. THEY SUFFER.
This is truly awfull.
Nothing calm, sparkly and diamondish about that.
This is diabolic fight with sulphur vapour pouring from under the heatspreaders.
Who said that? I was drinking last night but I wasn't that far gone. What you say here is what some of us originally thought. The GPU makes more sense to be fabbed at TSMC because they excel at low power, relatively speaking. Intel's commitment was solidified earlier this week with the announcement of their next gen GPUs. Whether Koduri is still involved in the project is another matter and question.
I wrote that the amount of suffering chips are subjected to is awfulll.What CPU do you use since they are all "truly awful?"
I wrote that the amount of suffering chips are subjected to is awfulll.
Building up pressure in the boilers is not a good way to win a race.
I dont think that squeezing any more frequency from the current process and making 13900KS in bulk is possible.
You must be new to this forum:I dont think that squeezing any more frequency from the current process and making 13900KS in bulk is possible.
You must be new to this forum:
- 6th gen - 4.2 Ghz
- 7th gen
- 8th gen
- 9th gen
- 10th gen ~ 5.2Ghz
Really?!You should have been able to OC Skylake to 4.8 at least.
Out of the four samples, one engineering sample had 4.7 GHz at 1.4V two engineering samples achieved 4.6 GHz at ~1.4V and the one retail sample had 4.5 GHz at 1.275 volts before declocking when it was running at 4.6 GHz / 1.4 volts.
Intel's just basically eaten any OC headroom recently to win stock benchmarks.
The timing is certainly going to be interesting. We know from an old leak that Intel originally planned to have Lion Cove + Skymont on N3 ready by end of this year, so really the question is what has slipped, and why.
Not just Pat the CEO saying the delay is bunk but Raichu here as well. Do wonder if ARL-S is planned to come out ASAP, like near Zen 5?
Was that to be taken at face value or that some part of the processor would be fabbed @ TSMC? Based on what limited info he probably had at the time I'm willing to let it slide. IDK who he is or if he's been correct in the past, but I wouldn't expect someone to know intricate details about the part of a multi chip processor months in advance of others.
Consolidated? I don't remember Lion Cove taping but I also don't pay attention to online discussion forums much.The timing is certainly going to be interesting. We know from an old leak that Intel originally planned to have Lion Cove + Skymont on N3 ready by end of this year, so really the question is what has slipped, and why.
It's good to see Paul Rudd being an overclocker.