It is a core. It's a gutted Zen 5 core where any parts that aren't strictly necessary for idle or low power usage are taken out.If I am understanding what is being discussed correctly, it is not a core at all - just like a filter on top of regular Zen 5 cores that can carry out a subset of instructions in a power efficient manner without waking up the full core.
Halo is the biggest part of the Z5 Strix lineup. Simple as that. It'll be expensive to buy, so they can afford to try things there and still run a profit. But Halo will probably turn out to be like what the 5800x3D was for AM4, a first run of an upcoming technology. They're using it as a first run for what's probably going to be commonplace on all Z6 mobile CPUs.If this is the case, why would this be exclusive to Halo? Perhaps because Zen 5C doesn't really benefit?
The reason was given by Rory Read at the time, it s because Intel was so in fear that AMD could capture the low power market that they started giving their Bay trails Atoms for free all while subsiding the OEMs, as stated by the AMD CEO there was nothing you could do with this, his formulae about this infamous and anti competitive scheme was "chips wrapped in $ bills", consequently AMD stopped developping those chips.The cats were built by a tiny team, that managed to massively outpetform expectations.
The reason cat core development ended was not about the market, it was that someone (iirc Samsung? Not sure and I don't have time to check it) poached the entire team whole from AMD. This was before the Zen success and when AMD had money issues.
It's literally called Zen5LP and it's a Zen5 derivative.No one said it has to be a Zen 5 derivative, right?
Zephyr small cores are located just beside Hurricane cores, like they're growing out of the big cores. Also, A10 b.l implementation was based on cluster migration were the P cores were the ones that did everything foreground. While the E cores handled idle, background, etc. But the key takeaway (imo) is this:
Emphasis on "custom designed performance controller to manage the CPU cores and migrate tasks between them"
Ok so the million dollar question: why did Apple give up on it immediately? Sounds like a pretty complete solution, but they gave up and went back to software scheduling anyway?This patent reminds me a lot of A10 Fusion, where the little cores ended up being almost transparent to software because if a workload was heavy enough it would transition from the little cores to the big cores, and primarily use those instead (not sure how that was determined by the OS/hardware, but I wouldn't be surprised if it relied upon the types of instructions run - like in that patent - or the workload duration).
A very hardware solution to the whole big.LITTLE scheduling problem. But Apple still ended up dropping it pretty quickly afterwards. It's why it was such a surprise to see AMD patent almost the same idea back then too.
Apple controls their own software stack. Microsoft cannot be trusted to make things work.Ok so the million dollar question: why did Apple give up on it immediately? Sounds like a pretty complete solution, but they gave up and went back to software scheduling anyway?
Aye but it seems like a hardware solution would be more final and complete and less taxing than a software one.Apple controls their own software stack. Microsoft cannot be trusted to make things work.
Cluster migration is expensive from a power standpoint. Unified scheduling from P to E cores also increases MT performance. Apple managed to stay with only 2P cores on the iPhone A series due to E cores being efficient while powerful enough to add performance on their own. Perks of controlling their own hardware and software stack.Ok so the million dollar question: why did Apple give up on it immediately? Sounds like a pretty complete solution, but they gave up and went back to software scheduling anyway?
*"Our Turin launch in the second half of the year"
Turin late 2024 launch and strix point H2 2024 launch. We are just 2 months away from H2. I thought they would have either said Q3 or Q4 at this point.
Introduced in Strix Halo? Or is it going to be in other products too?It's literally called Zen5LP and it's a Zen5 derivative.
Stix halo only, was mentioned earlier I this thread. Then likely (maybe) all zen 6 thereafter. In the segments it mattersIntroduced in Strix Halo? Or is it going to be in other products too?
Yeah STX1 and GNR are both Computex.Welp, Zen 5 at Computex confirmed. @adroc_thurston vindicated.
MC: "Coming out of that meeting, I just wanted to close my eyes, go to sleep, and then wake up and buy this thing. I want to be in the future, this thing is awesome and it's going be so great - I can't wait for it.""We’re witnessing a significant growth of local AI applications for both personal and business purposes."
In what way? He said Zen5 launch with absolute certainty should be in April, which did not happen. May 1 today, so it's now 100% confirmed he was incorrect.Welp, Zen 5 at Computex confirmed. @adroc_thurston vindicated.
Why the cope?He said Zen5 launch should be in April, which did not happen. May 1 today, so it's now 100% confirmed he was incorrect
Maybe as a lessons learned to not state that something is absolute certain and not up for discussion, when you're not sure about it. Words like probably, likely, etc would be more appropriate to use then.Why the cope?
It's completely unnecessary.
You're still gonna see Strix in August (even if it's like 2 laptops).
It's semis.Maybe as a lessons learned to not state that something is absolute certain and not up for discussion, when you're not sure about it. Words like probably, likely, etc would be more appropriate to use then.
Ok, so what do we make of that.It's semis.
Stuff is certain until it isn't.
Ok, so what do we make of that.
Can the promised average 40-45%+ perf increase core-to-core for Zen5 DT turn out to be only 20-30%, and can the $999 9950X price turn out to be e.g. $799?
This is wholly redundant cope again.Can the promised average 40-45%+ perf increase core-to-core for Zen5 DT turn out to be only 20-30%, and can the $999 9950X price turn out to be e.g. $799?
This again. You could have Dr Su answer a question about pricing or launch dates 6 months in advance and she'd potentially still be wrong.Ok, so what do we make of that.
Can the promised average 40-45%+ perf increase core-to-core for Zen5 DT turn out to be only 20-30%, and can the $999 9950X price turn out to be e.g. $799?