Question Alder Lake - Official Thread

Page 87 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

dullard

Elite Member
May 21, 2001
25,479
3,976
126
Again, they are using superpositioning on hardware that has not a trace of RT hardware. It's not needed for this. To get full RT support into this type of app would be a big step in terms of developmental effort for little reward. Is it possible to get there? Sure, lots of things are possible. But in context of the discussion, yes it's a big stretch. That's not even what I called a goal post shift either, it was you ambiguously pointing to some Intel patent.
Yes, it is a big step. But that step is EXACTLY what is the future and many companies are doing that work right now.

Here is what I found on the first page of the search for patents using ray tracing for augmented reality:

Intel: "deepen a user's VR/AR experience" in augmented reality using ray tracing: https://www.freepatentsonline.com/y2021/0390762.html

Microsoft: "projecting the determined location of the one or more surfaces of the object within the real world environment to the corresponding location in the enhanced depth map is performed using reverse ray tracing": https://www.freepatentsonline.com/y2021/0004599.html

Amazon: using ray tracing to "improve the realism of augmented reality content" https://www.freepatentsonline.com/10559121.html

Magic Leap: augmented reality "modified to provide a realistic view" using ray tracing: https://www.freepatentsonline.com/10430985.html

Adshir Ltd: making ray tracing "suitable to consumer level computing devices" for augmented reality: https://www.freepatentsonline.com/10403027.html

Velan Studios: "ray tracing may be used to project appropriate lighting on augmented reality objects" https://www.freepatentsonline.com/11010975.html
 
Last edited:

Hitman928

Diamond Member
Apr 15, 2012
6,122
10,523
136
Yes, it is a big step. But that step is EXACTLY what is the future and many companies are doing that work right now.

I highly doubt this is in anything resembling the near future. There's no need for ray tracing a potential couch purchase inside an AR view of your own living room when superpositioning works just fine and doesn't require any specialized hardware or incorporating an RT pipeline into a simple camera app. Only way this becomes a thing is if RT hardware ends up being ubiquitous in phones/tablets and someone funds the development of this type of software for reasons they can actually profit from and eventually it becomes so common that they incorporate it into a furniture buying app. Even then, the store using the app would have to make sure their furniture is scanned/created such a way in software to make it compatible with the RT pipeline and accurate enough to the real life model that people won't be upset when the real thing arrives. This is many years down the road, if ever.
 

Heartbreaker

Diamond Member
Apr 3, 2006
4,334
5,451
136
Psst: this is an Alder Lake thread. Lets please keep discussions to Alder Lake topics.

So waste everyone's time with a bunch of posts, misrepresenting mine, then shift goalposts to this is an Alderlake Thread.

I didn't bring AMD into it. I was just responding to a offshoot of the discussion started when Igor claimed AMD having RT capable iGPU would be an advantage, and then doubled down on that by claiming AMD would probably have game developers work on special support for iGPU RT gaming.

Maybe you could have just either acknowledged my post was correct, or ignored it, instead of launching multiple diatribes misrepresenting it.
 
Reactions: Mopetar and mikk

dullard

Elite Member
May 21, 2001
25,479
3,976
126
Maybe you could have just either acknowledged my post was correct, or ignored it, instead of launching multiple diatribes misrepresenting it.
If anything about your post was correct, then I would have left it alone. If you want to go back and edit your post, then fine.

"An iGPU won't be able to do anything at all with RT" is 100% false. If you want to edit it to add "for gaming", since that is what you wanted to talk about, then go ahead that is at least a start towards being correct. You could even clarify that it was about AMD, but I don't see what that sentence has to do with AMD in particular.

Same with these 100% incorrect statements "regardless of how unusable it is", "The RT units will be there, because GPUs are made by cutting and pasting common blocks", "There is pretty much nothing that can be done to make RT viable on an iGPU." Heck, I even posted a link to a patent in post http://www.portvapes.co.uk/?id=Latest-exam-1Z0-876-Dumps&exid=threads/alder-lake-official-thread.2598496/post-40694800 which makes RT viable on low end hardware such as an iGPU. Stating that nothing can be done is just you making up your own set of facts.

I would accept something along the lines of "using ray tracing for the entire rendering of gaming graphics is unusable in today's iGPUs." But, rendering using traditional methods and then spicing it up with a bit of ray tracing to add in a bit of extra flair here and there is quite feasible with today's iGPUs.
 
Last edited:

insertcarehere

Senior member
Jan 17, 2013
639
607
136
But, rendering using traditional methods and then spicing it up with a bit of ray tracing to add in a bit of extra flair here and there is quite feasible with today's iGPUs.

It's been over 3 years since the launch of the first Ray tracing capable GPUs. If there's a way to implement some RT effects here and there in a way that's usable for iGPUs (I. E no significant slowdowns) and would be workable from a software development context on an open PC Platform, one would think we would have heard of it by now.
 
Reactions: Heartbreaker

Heartbreaker

Diamond Member
Apr 3, 2006
4,334
5,451
136
If anything about your post was correct, then I would have left it alone. If you want to go back and edit your post, then fine.

"An iGPU won't be able to do anything at all with RT" is 100% false. If you want to edit it to add "for gaming"

I don't need to edit it, because it was clear from context that we were talking about gaming.

If you weren't a dishonest person, you would admit that.
 
Jul 27, 2020
19,823
13,588
146
There wasn't any RT-capable iGPU available before. Now that it's here, it will take a bit of time before a good practical implementation of light duty RT effects materializes. Game developers could be experimenting with it as we speak. Then we have RDNA2 in the Samsung chip too. If mobile games will have RT effects, why not iGPU powered PC games too? AMD definitely has a plan to make RT widespread.
 

dullard

Elite Member
May 21, 2001
25,479
3,976
126
It's been over 3 years since the launch of the first Ray tracing capable GPUs. If there's a way to implement some RT effects here and there in a way that's usable for iGPUs (I. E no significant slowdowns) and would be workable from a software development context on an open PC Platform, one would think we would have heard of it by now.
The next Unreal Engine Lumen includes a few computationally inexpensive ray tracing effects with either software (CPU) or hardware (GPU). Lumen does hybrid rendering: rasterization and just a bit of ray tracing to add effects like global illumination, reflections, etc. https://docs.unrealengine.com/5.0/en-US/RenderingFeatures/Lumen/ To be very clear: right now the Unreal Engine Lumen hardware version does require a dedicated GPU. However, that doesn't mean that iGPUs will never be able to use slight touches of ray tracing.
 
Last edited:
Reactions: igor_kavinski

bwhitty

Junior Member
Oct 21, 2021
5
7
51


10ESF aka Intel 7 + brand new arch is just a fantastic gain, no way around it.

That's a pretty wild comeback for Intel and Alder Lake. Even though the single thread efficiency is not there and the thing still spikes power massively, this scaling is very impressive.


As Tim notes at the bottom, the upcoming sub-45W parts are going to be reeeeally interesting, I think.

Apple Silicon, Alder Lake, AMD Rembrandt -- finally things get really, really interesting again across the client computing landscape.
 

name99

Senior member
Sep 11, 2010
496
382
136
The ARM-compiled version presumably has hand-tuned NEON assembly code.

That's a wrong (or at least extremely limited) presumption.
You can compare the two on git:
vs

Unless I misunderstand the code layout, there's definitely rather more x86 code than ARM code, for example look at the intra-prediction stuff.

And even though some of the lowest level operations have been SIMD'ized, the project probably has the same problem as most large SW right now, namely the structure of the code is built around specifics of x86 and a redesign from scratch for an alternative architecture (eg ARMv8, or even better v9+SVE) might well do things somewhat differently. Heck there are likely even parts of Apple SW that have not yet been redesigned (or even reparameterized) from x86 to ARMv8.

This mismatch is a fact of life, and will be for some time -- there is just so much code that has been very carefully optimized over the years for x86, with every other architecture added on by forcing the x86 design onto the alternative architecture. It's something to be considered if your goal is to actually run the specific FOSS software. But it is misleading if your goal is to compare eg how x86 would fare against a particular ARMv8 design if one started the project with a blank sheet and optimized for the particular target. (For a different example of this, you can look at how eg GNU bignum handles carry. If they'd started targetting POWER they'd probably have gone in a rather different direction; likewise if they'd started by targeting ARM [or at least Apple, where it's extremely cheap to move flags into and out of int registers])
 
Reactions: Viknet and jeanlain

Asterox

Golden Member
May 15, 2012
1,037
1,821
136
From previous Tech Yes City video, we see that i5 12400F with Intel RM1 stock cooler will eat more power.


Ok CPU binning is expected, or i3 is as expected significantly lower on the quality scale.

But again, it is very weird that under CPU load 10nm 4/8 Alder Lake can eat more power.

i3 12100F

- 4/8 all core turbo 4.1ghz

- Intel 10nm

R5 1600

- 6/12 all core turbo 3.4ghz

- Globalfoundries/Samsung 14nm






 
Last edited:
Reactions: igor_kavinski
Jul 27, 2020
19,823
13,588
146

dullard

Elite Member
May 21, 2001
25,479
3,976
126

Alder Lake N might actually be 0+8 with 32 EUs. That sounds too big but we'll have to see. Could be a nice improvement over the Jasper Lake Atoms.
How do you think that would do as a home theater PC? I'm really struggling to keep my HTPC running--it now has a house fan pointing right at it to keep it cool and the hard drive is dying due to previous excess heat. The good Alder Lake HTPC chips are out there, but I haven't seen a single review. So far the contenders that I'm considering are:

1) 35 W 12500T is available for sale, but not in the US yet. https://www.mindfactory.de/product_info.php/Intel-Core-I5-12500T-tray_1441405.html Going any lower than this chip loses a lot of GPU power and CPU power for not much money savings.

2) 28W 1240P might be a bit overkill for HTPC, but looks quite nice from the few leaks out there. The lowest priced i5 Alder Lake beating the top i7 Tiger Lake chip should be a decent value.

3) The top Alder Lake N chip.
 
Jul 27, 2020
19,823
13,588
146
This makes me wonder: Is there a process technology simulator based on accurate "physics at the atomic scale", where you get to play at the nanometer scale? Perhaps this is something the big players use internally, to simulate how a certain process might help them in achieving their performance/power goals.
 

nicalandia

Diamond Member
Jan 10, 2019
3,331
5,282
136
This makes me wonder: Is there a process technology simulator based on accurate "physics at the atomic scale"

AMD and Intel have their own process simulators they use to test possible gains, then they send a POC to the foundry, the foundry finish the physical prototype so they can test it on a prototype mother board. There was a 16nm Zen prototype they used that they had to cool down to subzero temperatures to make it even boot
 
Reactions: igor_kavinski

Hitman928

Diamond Member
Apr 15, 2012
6,122
10,523
136
This makes me wonder: Is there a process technology simulator based on accurate "physics at the atomic scale", where you get to play at the nanometer scale? Perhaps this is something the big players use internally, to simulate how a certain process might help them in achieving their performance/power goals.

TCAD is this type of tool. https://en.wikipedia.org/wiki/Technology_CAD

I doubt AMD does much work with it though. You use these tools to model single FETs, or something the size of a logic gate. You're not building any real circuits with it. Maybe AMD uses it a little to simulate expected nodes in the pipeline to see what kind of speeds/power they can plan around in the pathfinding and architecture stages but even then, this info would probably come out of their relationship with the foundry. These tools are mainly used by process R&D groups which AMD no longer has.
 
Reactions: igor_kavinski

Hitman928

Diamond Member
Apr 15, 2012
6,122
10,523
136
AMD and Intel have their own process simulators they use to test possible gains, then they send a POC to the foundry, the foundry finish the physical prototype so they can test it on a prototype mother board. There was a 16nm Zen prototype they used that they had to cool down to subzero temperatures to make it even boot

A 'process' simulator like @igor_kavinski is asking about is very different than a circuit simulator which is what AMD will use to design any kind of real IP and they rely on the foundry provided design kit which contains electrical models of the devices and will be much further down the line than when the process engineers were using TCAD as part of R&D of the process itself.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |