Saturday, January 20th 2024

Detailed Intel Arrow Lake-S Platform Specifications Leaked, Confirms Native Thunderbolt 4 Support

Courtesy of X/Twitter user @yuuki_ans, we now have what should be very detailed information on Intel's next generation consumer desktop platform, assuming the leaked information is real. The leaker not only provided confirmation on the CPU specs of the Arrow Lake-S desktop CPUs, which will feature up to an 8+16+1 core configuration. However, it appears that it's not all smooth sailing for Intel to get Arrow Lake-S up and running, as a note points out that the pre-alpha hardware has the performance cores disabled due to a hardware bug that is expected to be fixed in a future hardware revision. We can also see that the official memory support is DDR5-6400 from the block diagram, which is quite a jump from DDR5-5600 which is what the current 14th gen CPUs officially support.

That said, the rest of the documentation shared is very detailed and provides us with a ton of details in terms of the various platform interfaces we can expect. For starters, the Arrow Lake-S CPUs will feature native Thunderbolt 4/USB4 support (once again an odd mistake here stating USB4.0), as well as DisplayPort 2.0 (UHBR20 only) and HDMI 2.1 support. The CPU is said to deliver 24 PCIe lanes, of which 16 are PCIe 5.0 lanes for the GPU and the remaining eight are for NVMe SSDs, with half being PCIe 5.0 and half PCIe 4.0.
The chipset which is expected to carry the 800-series naming will connect via a DMI Gen 4 interface with up to eight lanes to the CPU. Here we can apparently expect a further 24 PCIe 4.0 lanes, up to 10 USB 3.2 Gen 2 (10 Gbps) ports, of which eight can be combined to offer USB 3.2 Gen 2x2 (20 Gbps) ports instead. That said, there are some oddities in this leaked information that puts up some red flags. The above mentioned IA Cores and USB4.0 mistakes are odd and there's also a suggestion that the CPU should feature eight SATA 3.0 ports (muxed with PCIe) and the chipset featuring a further eight (also muxed). Furthermore WiFi support is only listed at 802.11ax R2 or WiFi 6E as it's now better known as, but that there will be Bluetooth 6 support in the same module.

Now keep in mind that we're looking at the specs for an Intel test platform here, so we might just be looking at a sloppy engineer that threw this document together to be able to send along some specs to Intel's partners that are getting boards to start to work on their UEFI implementations and board designs. That said, there are a few too many oddities in this leak to be able to say that the features are set in stone or even that it's a real leak, but it does reflect what is already known about Arrow Lake-S. As such, take this for what it is, namely an interesting look into what might be.
Sources: @yuuki_ans, via VideoCardz
Add your own comment

35 Comments on Detailed Intel Arrow Lake-S Platform Specifications Leaked, Confirms Native Thunderbolt 4 Support

#2
Dristun
Bet the cheapest board with a couple of TB4 ports will still be $400+, duh
Posted on Reply
#3
Onyx Turbine
What does the snip mean with 8 cores 16. Is it a max 8p core and 16 e core config? 125 watt tdp? what does the +1 mean?
Posted on Reply
#4
TheLostSwede
News Editor
ChippendaleWhat does the snip mean with 8 cores 16. Is it a max 8p core and 16 e core config? 125 watt tdp? what does the +1 mean?
Yes, that. The additional core is another E core for system management from what I understand.
DristunBet the cheapest board with a couple of TB4 ports will still be $400+, duh
Tree fiddy, the other fiddy ends up on the CPU side.
Posted on Reply
#5
Onyx Turbine
TheLostSwedeYes, that. The additional core is another E core for system management from what I understand.


Tree fiddy, the other fiddy ends up on the CPU side.
ok interesting, although i was under the presumption there would be also a 65 watt tdp?
Posted on Reply
#6
TheLostSwede
News Editor
Chippendaleok interesting, although i was under the presumption there would be also a 65 watt tdp?
I'm sure there will be a range of parts. Keep in mind that this is just some kind of early engineer sample.
Posted on Reply
#7
kondamin
how old is the document?
Posted on Reply
#8
TheLostSwede
News Editor
kondaminhow old is the document?
No idea, but it was posted today on Twitter, but taken down later.
Posted on Reply
#9
qcmadness
To support 2 ports of Thunderbolt 4, should not be there additional PCIe lanes (two x4) present?
Posted on Reply
#10
TheLostSwede
News Editor
qcmadnessTo support 2 ports of Thunderbolt 4, should not be there additional PCIe lanes (two x4) present?
Why? They're integrated into the CPU die, not relying on an external chip.
Intel already does this with some mobile CPU SKUs.
Posted on Reply
#11
qcmadness
TheLostSwedeWhy? They're integrated into the CPU die, not relying on an external chip.
Intel already does this with some mobile CPU SKUs.
I mean the two ports can support 2 x4 lanes externally.
Posted on Reply
#12
TheLostSwede
News Editor
qcmadnessI mean the two ports can support 2 x4 lanes externally.
Yes? But what does this have to do with PCIe lanes in the CPU/on the motherboard?
It's straight Thunderbolt 4/USB4 from the CPU to the physical ports, with some signal redrivers/retimers on the board.
Posted on Reply
#13
DavidC1
ChippendaleWhat does the snip mean with 8 cores 16. Is it a max 8p core and 16 e core config? 125 watt tdp? what does the +1 mean?
+1 means GT1, meaning the base 64EU configuration. Yes 8+16 means 8P and 16E.
The leaker not only provided confirmation on the CPU specs of the Arrow Lake-S desktop CPUs, which will feature up to an 8+16+1 core configuration, plus eight "IA Cores" which was presumably meant to read AI cores in the document that the screenshots are from.
No you dumbo author. IA = Intel Architecture. That's what's referred to their own cores. Was THIS article written by AI, or was the brain hijacked by it?
Posted on Reply
#14
Onyx Turbine
it would be nice to see a 8p core chip which is reasonably priced around the 225 dollar range,
that would be a keeper
Posted on Reply
#15
Wirko
PCIe bifurcation! The existing Intel and AMD consumer platforms are painfully deficient in that regard. Hopefully that's about to improve.
Posted on Reply
#16
efikkan
If this is genuine (which it may or may not be), the first thing that stands out is that the P-cores are apparently defective, which is no small deal considering the core feature is missing.
If this is the case, this would be a very early engineering sample, and probably something only distributed to very few partners, which in turn wouldn't make it hard to find out who leaked this.
Posted on Reply
#17
DavidC1
efikkanIf this is the case, this would be a very early engineering sample, and probably something only distributed to very few partners, which in turn wouldn't make it hard to find out who leaked this.
That's why most legitimate leakers get hired by the company. Ashraf Eassa was a notable leaker for Intel products. A year or so later Intel hired him. No more Intel leaks.
Posted on Reply
#18
Broken Processor
I wish Intel would release a gaming CPU I'm not paying for e-cores I'm not using and yes I know window will use e-cores when a game is using p-cores but I'd rather have the e-core space populated with p-cores instead. But now I suppose we have to also contend with AI core bollox, can't wait for that bubble to burst and it will since the commercial AI we have now is nothing even remotely like proper AI. Sorry for AI tangent.
Posted on Reply
#19
kondamin
TheLostSwedeNo idea, but it was posted today on Twitter, but taken down later.
I hope for intels sake it’s old.
half the package not working doesn’t bode well for a 2024 release
Posted on Reply
#20
TheLostSwede
News Editor
WirkoPCIe bifurcation! The existing Intel and AMD consumer platforms are painfully deficient in that regard. Hopefully that's about to improve.
Sorry? It hasn't been an issue on AMD since they launched Ryzen. It does depend on UEFI support though.
Posted on Reply
#21
Tek-Check
TheLostSwedeAnd for those interested, Intel Hayden Bridge is already available from Intel at US$3 a pop and is a signal redriver/retimer.
The reason it connects to the CPU is due to the native Thunderbolt 4/USB4 support.
www.intel.com/content/www/us/en/products/sku/211299/intel-jhl9040r-thunderbolt-4-retimer/specifications.html
I have seen one single product with this retimer chip - Asrock's mini-PC. AIBs don't seem keen on it, do they?
WirkoPCIe bifurcation! The existing Intel and AMD consumer platforms are painfully deficient in that regard. Hopefully that's about to improve.
It depends which chipset you are talking about. Highest chipsets are quite flexible with bifurcarion, from x1 all the way to x8.
Posted on Reply
#22
TheLostSwede
News Editor
kondaminI hope for intels sake it’s old.
half the package not working doesn’t bode well for a 2024 release
Impossible to tell I'm afraid, as all the dates that a PDF like this normally has, aren't included as you can see.
Tek-CheckI have seen one single product with this retimer chip - Asrock's mini-PC. AIBs don't seem keen on it, do they?
I wasn't even aware of that part, but I haven't exactly been looking either. Steep price, that's most likely why companies are using cheaper third party options. My understanding is that most things like that are around US$1.
Posted on Reply
#23
Wirko
TheLostSwedeSorry? It hasn't been an issue on AMD since they launched Ryzen. It does depend on UEFI support though.
At the very least, the PCIe 5.0 x4 interfaces on the CPUs should support splitting to x2 + x2. Samsung's 990 Evo SSD, if it becomes a reality, is an indication that this is going to happen.
Posted on Reply
#24
TheLostSwede
News Editor
WirkoAt the very least, the PCIe 5.0 x4 interfaces on the CPUs should support splitting to x2 + x2. Samsung's 990 Evo SSD, if it becomes a reality, is an indication that this is going to happen.
They do. But no-one has implemented it slot wise, as the motherboard makers most likely don't think anyone is going to want to lower the speed of their PCIe 5.0 x4 drives in favour of being able to have an additional drive.
PCIe obviously allows for down to x1 in a x16 slot, so if you want to install some janky M.2 adapters to your M.2 slots to split four lanes into two, go ahead, it will work, at least on AMD.
And before you ask, yes, someone made one already.
www.amfeltec.com/flexible-minipci-express-2way-splitter/
Posted on Reply
#25
Wirko
TheLostSwedeThey do. But no-one has implemented it slot wise, as the motherboard makers most likely don't think anyone is going to want to lower the speed of their PCIe 5.0 x4 drives in favour of being able to have an additional drive.
PCIe obviously allows for down to x1 in a x16 slot, so if you want to install some janky M.2 adapters to your M.2 slots to split four lanes into two, go ahead, it will work, at least on AMD.
Epycs, Threadrippers and Xeons are all really good when it comes to PCIe 5.0 lane splitting, with Xeons being a bit inferior. But consumer stuff? Intel has 8+8+4 and AMD has 8+8+4+4 if we count all PCIe 5.0 lanes available in the system. The 5.0 x4 ports don't allow further splitting (I've never seen it documented anywhere) and the 5.0 x8 ports (two halves of the PEG bus) may allow splitting to x4. Neither Intel nor AMD mandate that, so the support depends on the motherboard manufacturers.

And I think bifurcation is becoming/will become more appealing now that we have 5.0. "I don't need the ultimate speed, I need the lanes!" is an opinion I've often seen expressed here at TPU forums. Those who need the lanes actually need the ability to connect many devices, SSDs, graphics cards, network cards, basically to squeeze out workstation-level connectivity from a consumer desktop system.

The splitter you linked to ... it's unclear if it's a passive splitter or it has a PCIe bridge chip onboard. The price difference is huge if we're talking PCIe 5.0 or 4.0.
Posted on Reply
Add your own comment
May 20th, 2024 20:48 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts