dylan522p

6 ms·

1.
by dylan522p 4y ago
The article does answer the question, in detail. What you're reading is just the first half is freely available. The 2nd half is on the same web page and for subscribers.
2.
by dylan522p 4y ago
They use some of this too btw. Also wavelength level routing happens with breakout cables from ToR to Compute.
3.
by dylan522p 4y ago
OPC? It'd be great to talk about your area, because I bet most don't know what you do, and maybe I can make it an entertaining read.
4.
by dylan522p 4y ago
No TPUs are getting better architecturally
5.
by dylan522p 4y ago
Your math is completely wrong dude. It's 2000ms per token not for the whole query. Hardware utilization rates and MFU are not the same thing, you forgot the latter. You are pretending its perfectly parelelized on 1 GPU too. I use 8x GP
6.
by dylan522p 4y ago
> This article is based on wild speculations of how much things cost Huh it literally uses real throughput figures > doesn’t account for ways to make things cheaper over time It does in the subscriber section and it says it does say t
7.
by dylan522p 4y ago
- Google qps is closer to 100k then 320k [1] That number is wrong. I have a number from googler, not livestats which cannot have google internal data - Not every query has to run on LLM. Probably only 10% would benefit from it Agree, i have
8.
by dylan522p 4y ago
cutlass is faster actually, in most cases where cutlass supports same stuff.
9.
by dylan522p 4y ago
Thanks! Ya, the archive link is useless, it only captures the free part, and I think I am very generous with what I keep in the free section on the ad-free website.
10.
Caliptra – First Open-Source Silicon Going into All Datacenter Chips(semianalysis.com)
2 points by dylan522p 4y ago|0 comments
11.
by dylan522p 4y ago
Tsmc 7nm doesn't use EUV and is fantastic
12.
by dylan522p 4y ago
Author here. What do you mean die shot bleeding?
13.
by dylan522p 4y ago
M1 2020, M2 2022?
14.
by dylan522p 4y ago
It has noticed. Look at Apple architects, validation, layout, etc engineers moving to Nuvia + Rivos + Google + Amazon + Microsoft + Meta + Intel + Nvidia + AMD + Apple + Qualcomm. It's there.
15.
by dylan522p 4y ago
It's at the end of every article lol
16.
by dylan522p 4y ago
> the unsubstantiated BS that Samsung's chip manufacturing is a disaster. They lost Qualcomm, Nvidia, and Cisco for the next generation. Are you disputing this fact? They did not fully ramp 1Z or 1 Alpha dram nodes. Are you disputin
17.
by dylan522p 4y ago
I have been sent C&D letters in the past by Arm, and even sued by others. I have resources.
18.
by dylan522p 4y ago
If you don't want to believe it. Go ahead. The major claims are true, Qualcomm moving away. Nvidia moving away. DRAM node ramps being pitiful. DRAM engineering efforts come directly from a source there. The cultural issues being the ca
19.
by dylan522p 4y ago
MediaTek is killing it. They hired a lot of people from TSMC's process development kit teams which vastly improved their capabilities in integrating IP. They were the first to develop AV1 decode, their modems are pretty decent. They ev
20.
by dylan522p 4y ago
Most folks blame the culture shift on Paul actually. He was the cause of it all. Brian being a horrible symptom of what he created, and Brian definitely propagated it far.
21.
by dylan522p 4y ago
If anything I am making the case that TSMC is reigning supreme but okay, believe your conspiracy
22.
by dylan522p 4y ago
Good thing the article even says Samsung is a well oiled machine in many areas outside of the ones with issues discussed in this article. Almost like large companies can have varrying success and culture.
23.
by dylan522p 4y ago
It's at the bottom of every article. I have to disclose given my relationship with various funds. I am under no impression that my articles can move fucking Samsung's stock. That's hilarious you think I could profit off the m
24.
by dylan522p 5y ago
My original article you are talking about explicitly stated heterogenous compute is the future. It talks about transistor budgets going to LLC, GPU, media, and ISP. Much of the CPU efficiency gain comes from that doubled LLC size.
25.
by dylan522p 5y ago
Sophie's analysis contemplates a set unit volume and fixed costs related to design and verification and tape out. Mine is on pure cost/transistor based on wafer costs as a player like Apple has the units to spread fixed costs over
26.
by dylan522p 5y ago
Look at the in depth analysis I did of the die shot and SOC floorplan. Or even read the article I originally wrote I was right on the CPU architectural gains. Apple pushed clockspeed up slightly, but IPC was less than 5%. https:/&#x2F
27.
by dylan522p 5y ago
Why no on the BOM increase? That is what the switch to LP5 and A16 would be.
28.
by dylan522p 5y ago
The IPC gains were what I specifically wrote about and those turned out to be true with <5% IPC gains on the Avalanche core CPU
29.
by dylan522p 5y ago
They are allowed to sell chips outside of china with Arm IP licensed from Arm China
30.
by dylan522p 5y ago
The last part of the article was not here last year. The last part and images are from an event they held recently. Notice the "Before we get to the event they held and the significance of it, let’s do a recap." He refuses to leav

More ›