NUCs are a niche of B&M shops - you really don't need mobility at a front desk. However, you need *some* mobility at nearly any other position.
* Being at home office for a day -> laptop
* Attending a meeting -> laptop
* Solving a cross-team project issues -> laptop
* Working from an office...
Pre-launch BD dividied the forum crowd to two groups:
* the hype train ppl - "all new architecture", "8 cores", "SMT killed by cluster", "FMA4!"
* the IPC fail ppl - "2 ALUs", "0.5 FPU", "16k L1D lmao"
On the other hand, Zen 5 has been hyped to the sky since that weird remark in that interview...
Zen 5 is somehow going to be somehow presented at Computex. That's June 3rd. Yet, nobody knows the real release date or availability.
Retail AM5 BIOSes are ready but samples are not leaking apart from that Strix Point Geekbench. Either this time there are some super advanced restrictions...
What has a server Zen 5 has to do with AMD Q1 results?
Turin comes in H2. There are going to be "30% more designs than Genoa" - at some point in 2024-2025. Ugh
Intel will happily keep selling their Emeralds through OEMs.
The linked article lists AMD 9554 averaging at 227W and Intel 8592+ at 289W. Peak power figures are 369W and 434W. That's not 50% in my books.
Anyways, Intel sold tons of Skylake servers vs Rome. OEMs do not care. Now Intel is way closer than it was in Skylake times.
Technical superiority vs Emerald Rapids? They are more or less the same, unless you aim for those super premiums - 96-128c. Also there are non-technical aspects like - Intel price war or AMD's OEM relations have always been bad.
Strix won't shake anything if they can't push it to the OEMs...
The problem is where is no universal "Linux" as a platform. There are completely different kernel + user space setups and their versions. Besides, anyone can configure and build their own kernel - optimize for latency, throughput, powersaving, or something else? Also the compiler flags can...
Judging by the usual AMD mobile market performance, there are going to be 1-2 "exclusive" (think overpriced weird) laptop models featuring Strix Halo.
However, it is great it is still on the roadmap and it might see the light of day.
This is strange. The open source ecosystem surrounding Linux is a nearly endless source of tinkering, tuning, hacking, and - most importantly - a way to "know how things work under the hood". Discovering new things and understanding how technology really works can be an exciting experience. No...
Well, how different is this new MLID's email slide from those older MLID's Zen 5-related ones? I haven't checked them side-by-size but they both present nearly identical info, right?
Nah, the rumor mill presented real killer IPC gains for Zen 4 back in the days. Nobody expected a ~11% IPC gain with a much steeper frequency gain. Relevant MLID stated: "above 20% IPC increase over Zen 3".
Sorry, you must be new here.
"My sources got old data". "My source tested an old firmware". "My IPC figure didnn't count games". "My IPC figure got no AVX-512".
The same clowns pushed the Zen 4 killer IPC or RDNA3 brutal leakz. Navi 33 having 4+k stream processors, being a RV770 reborn, with a...
The green ppl simply deliver. They keep showing their laughably misleading slides with 2x/3x improvements but... they deliver.
RTG went through a rework after Raja got finally kicked out. During RDNA2 dev they shifted a lot of ppl from CPUs to RTG. But apparently they delivered RDNA3 after...
Oh...
They couldn't do a postmortem using N31/N33 any of pre-prod/release samples. Right.
It's like they did not know they got a problem until like a month to the N31 release. Right.
N31 got quickly showed out of the door with utterly rushed drivers/firmware - yet there was no time for a...
RDNA3 timeline:
* the ultimate nVidia-beater - Navi 31 - 15k shaders => over 90TFLOPS!
* still utter devastation - Navi 31 - 12288 shaders is over 2x of Navi 21's 5120!
* wait for the RX7900 respin - Navi 31 is broken - revision A0, lol!
* no respin, but RDNA3 is fixed in later designs like Navi...
Try to think out of your power-user bubble.
Think about the average PC workload. The wast majority of PCs is used for office work. These days it usually means interacting with one or more webapps in a browser. Plus Electron(-ish) Teams/Outlook if the corp goes the MS way.
Sadly, JS has been the daily driver for like 90+% of *standard* workloads.
* browsers - ranging from Anandtech formus, FB chat, YT videos, Google Docs, or even fully embedded VSCode IDE or Excel
* Electron-based apps - Slack, Teams, or Discord IMs; next-gen Outlook; VSCode or Sublime IDEs
*...
For Zen 4 initial impression add the power/temp hike along with the not so popular "freq gains > IPC gains" formula.
TBH I still don't know how much important the DIY desktop market is besides its marketing influence. OEMs markets are way larger, right?
AMD was working on Zen 5 back in April 2018. That's over 6 years. So far so good.
Just note the 40% core-to-core SPECint figure *might* come from a 64-128c server part.
That'd be a completely different realm compared to 8-16c desktop. The scope of many-core server is quite wild. Zen 5 features...
What do you mean "finally"? As AMD stated, Zen release cadence was 12-18 months and Zen 4 had been delayed due COVID and CXL 1.1. This means Zen 5 launches in April. AMIRITE?
AmpereOne has been reportedly sampling since May 2022, launched in May 2023, and agot dopted by Oracle in Sep 2023. Yet, AFAIK there are no 3rd party reviews available nor is the Oracle A2 instance.
AMD stated they are ready to adopt ARM at any time it makes sense.
ZenX branding will surely...
Official public support of Vermeer on AM4 arrived ~month prior its launch IIRC. There surely are dev platforms with dev BIOSes supporting Zen 5.
So as always with an existing platform - we first need some BIOS activity (MCExtractor & co.) weeks prior launch.
MLID's apparently been able to present genuine AMD NDA-protected slides for some time. So my advice is to 100% disregard all his written comments and focus just on the 3rd party original slides.
Zen 5 original slides so far has been confirmed 100% true.
The Zen 4 enablement is not indicative since it wasn't really different form the rest of Family 19h from the machine PoV. OTOH the Zen 3 situation was similar to the Family 1Ah/Zen 5 one. Yet the Zen 3 GCC patches were published early Dec 2020 - a month after the launch.
Seeing both binutils...
Lower clocks are to be expected compared to Raptor-R - the question is how much lower. The RAM latency is probably gonna suffer simply because of using a disaggregated design. Absence of HT is still a bit murky given we still have only early samples of Arrow/Lunar.
Usually, lower frequency is...
Major changes always bring much more risk than incremental ones. It's a question of risk management. The Zen 5 development seems to be less turbulent than Bulldozer times (released 4 years after the first failed 45nm attempt).
Anyways, is Zen 5 such a major change like the 10h->Bulldozer...
Btw the recent JPR report shows how the PC market segmentation works:
70% of PC CPUs are notebooks. Only 30% are desktops - OEM & DIY. This means AMD's focus point should really be APUs and OEM channels.
In high-GDP coutries (US, West EU, China?) Macs are usually used by developers for a simple reason - Windows don't feature Bash (WSL not counted). The rest of the world uses a mix of Mac/Linux/Windows.
Windows is drifting towards entertainment systems filled with ads. The interaction with...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.