PC Perspective Podcast #399 – 05/11/2016
Join us this week as we discuss the GTX 1080 Launch, UWP Updates, DOOM Vulkan Patch, Kaby Lake Leaks, ASUS ROG STRIX X99, and more!
You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.
The URL for the podcast is: https://pcper.com/podcast – Share with your friends!
- iTunes – Subscribe to the podcast directly through the iTunes Store (audio only)
- Google Play – Subscribe to our audio podcast directly through Google Play!
- RSS – Subscribe through your regular RSS reader (audio only)
- MP3 – Direct download link to the MP3 file
This episode of the PC Perspective Podcast is sponsored by Casper!
Hosts: Ryan Shrout, Jeremy Hellstrom, Josh Walrath, and Allyn Malventano
Program length: 1:03:21
-
Week in Review:
-
-
AD BREAK
-
News items of interest:
-
Hardware/Software Picks of the Week
-
Jeremy: Canadians are getting fibre!
-
Allyn: Old retro gaming history from PlayValue
-
Sebastian: Turn your Nexus into a Wi-Fi hotspot!!
-
Closing/outro
Subscribe to the PC Perspective YouTube Channel for more videos, reviews and podcasts!!
It’s never been a better time
It’s never been a better time to be a gaymer, a PC gaymer. Why isn’t PC per doing a Doom benchmark article or series. The hits you’d get would be off the charts.
Get cracking, there’s tons to explain with Vulkan to explain to the unwashed masses, AMD vs nVidia, the upcoming 1080 (which you better have doom benches for, I don’t wan’t to see Metro in there).
So get with it, TotalBiscuit gets sever hundred thousand views for simple port reports on new games, this type of info should be PC Per’s bread and butter.
So let’s stop promoting Patreon and get cracking. Your article on AMDs long slog with support was great but let’s apply that hard work to yourselves.
Let’s make PC Per great again.
Bethesda didn’t send out
Bethesda didn’t send out advanced copies of Doom, how could they do a real review unless it was based on the beta?
Here this gamers! AMD needs
Here this gamers! AMD needs to get more of that HPC/Server/workstation business, so hope for more of AMD’s Zen/Vega HPC/server/workstation business because that’s where the big markups are, and that produces the revenues for the most R&D! That’s why Nvidia has all that cash to get get its gaming SKUs to market sooner! The consumer gaming market alone does not produce enough revenues for AMD, so hope of more Zen/Vega HPC/server/workstation business to allow AMD to do more of the Nvidia like things(Getting its SKUs to market sooner), AMDs asynchronous compute is already damn good and ready for the VR gaming market.
Nvidia good work on getting that GPU thread process granularity down from the block level to the instruction level on you Pascal GPU Micro-Architecture. So those GPU threads can be swapped out more efficiently and your GPU execution resources can be better utilized.
Both Nvidia and AMD start getting even more of the traditionally CPU like instruction functionality into your respective GPUs’ Micro-architectures so the VR games can get rid of that nasty and cookie-tossing side effects and such, because latency is a very sickening thing for VR gaming. More of the non gaming graphics compute done on the GPU, in addition to the usual graphics compute reduces that CPU to GPU over PCIe latency inducing encode/decoding over the PCI protocol to a minimum! Now if AMD and Nvidia will just start adding some on GPU specialized Ray Tracing units on their GPU SKUs things will be even better for more realistic graphics and more natural lighting/shadow effects, for both gaming graphics and non gaming graphics workloads. Imagination Technologies PowerVR mobile graphics offers GPU options with specialized in the GPU’s hardware Ray Tracing units, so the mobile markets are getting GPU hardware Ray Tracing units before the discrete market is getting them!
And MORE Vulkan/Linux based gaming reviews, please keep up with more of the Steam OS ecosystem benchmarking when the Vulkan/Linux(Steam OS) benchmarking starts coming online!
I really look forward to Asus
I really look forward to Asus X99 Strix. Sweet! Broadwell-E baby!
“AMD to give an ‘Inside look
“AMD to give an ‘Inside look at Polaris’ on May 18th”
http://videocardz.com/59923/amd-to-give-inside-look-at-polaris-on-may-18th
Yay more asinine “Anonymous”
Yay more asinine “Anonymous” posts that have no relevance to anything on this site.
Mindless Gaming GITs! The
Mindless Gaming GITs! The incestious spawn of the world! Isn’t your SMOM’s stripping job paying for that internet access!
“AMD Unleashes Initial AMDGPU
“AMD Unleashes Initial AMDGPU Driver Support For GCN 1.0 / Southern Islands GPUs”
https://www.phoronix.com/scan.php?page=news_item&px=AMDGPU-SI-Experimental-Code
“AMD Delivers OverDrive Overclocking Support For AMDGPU DRM Driver”
https://www.phoronix.com/scan.php?page=news_item&px=AMDGPU-OverDrive-Support
“AMD Confirms Sony
“AMD Confirms Sony PlayStation Neo Based on 14nm CPU and Polaris?”
He, he he!
http://vrworld.com/2016/05/11/amd-confirms-sony-playstation-neo-based-zen-polaris/
“The new 14nm FinFET APU
“The new 14nm FinFET APU consists out of eight x86 ‘Zen Lite’ LP cores at 2.1 GHz (they’re not Jaguar cores, as previously rumored) and a Polaris GPU, operating on 15-20% faster clock than the original PS4.”
That will be very interesting if true. Hopefully, we will see such APUs in mobile parts. Although, it is unclear what a “Zen lite” would actuall be. It would probably still be plenty powerful enough for a laptop if it has 8 cores of threads.
Or 8 threads, that is.
Or 8 threads, that is.
Interesting on the rumored
Interesting on the rumored Zen Lite cores!
Remember when AMD took the
Remember when AMD took the 28nm Carrizo CPU core design and used it’s GPU high density design libraries to layout the design of the Carrizo CPU cores, well that reduced in the planar dimension by 30% the amount of die space occupied by a Carrizo CPU core, without any fab process node shrink.
Now using the GPU design libraries in the design layout for a CPU core results in a denser circuit packing, but it also reduces the ability of the cores to be clocked as high as a CPU designed using the low density high performance design libraries normally used in desktop CPU core designs. So AMD if they used high density GPU design libraries for the PlayStation Neo’s CPU cores Zen(Lite) design will get that 30% extra space savings at 14nm(in addition to the 14nm node shrink’s space savings) and have even more room for more GPU assets, if AMD is using a single monolithic die based APU!
Now if the Design uses HBM, and it is of an APU on an interposer design, things can be very different and that PlayStation Neo could be very powerful with a Interposer etched wide CPU die to a GPU(larger) die connection fabric that could rival the effective bandwidth of any CPU to GPU over PCIe connection’s bandwidth, or NVLink’s effective bandwidth, just look at the GPU to HBM effective bandwidth at clock rates 7 times slower that of GDDR5! So if AMD has gone with a full APU on an interposer option then there will be much more flexibility. This APU on an interposer design will probably first be used in the HPC/workstation/server markets where the markups are plenty high to pay for any new and currently more costly HBM2 based designs.
Something else to make note of, there is nothing stopping AMD/JEDEC from having a new HBM standard that allows for a more than 1024 bit wide channel/s to each HBM stack, to allow for higher density HBM stacks that allow for the same effective bandwidth to 2 higher memory density HBM stacks instead of 4 HBM stacks. It’s just a matter of costs versus interposer space savings. But something like this would allow for more space for GPU/other assets over the HBM stack’s interposer footprint. It all depends on the largest size of interposer dimensions available.
I’m thinking that this will be of a single monolithic die based APU, similar to the other console SKUs, But AMD could still make a CPU/GPU monolithic die and wire that die up to some HBM die stacks, even 2 HBM2 die stacks at the cost of some effective bandwidth, but that depends on the bandwidth needs of the GPU and the Zen(Lite) cores on the PlayStation Neo! They still may go with GDDR5X at the cost of more space/power required dew to HBM/HBM2 availability.
“Can Open Source Hardware
“Can Open Source Hardware Crack Semiconductor Industry Economics?” (1)
RISC-V is an open source ISA.
“The open source RISC-V effort, with designs that support up to 128-bit memory addressing, has wide potential and is fully open source, and now has support from more recent members of the organization, including Google, HPE, Lattice Semiconductor, Oracle, and others as well as a story of major progress with the 64-bit Shakti processor out of IIT Madras, which now has plans for at least six microprocessor designs along with fabrics and an accelerator chip.” (1)
(1)
http://www.nextplatform.com/2016/05/16/can-open-source-hardware-crack-semiconductor-industry-economics/
Want to see just where MOST
Want to see just where MOST of that Samsung/HBM2 is going to be going on Nvidia’s side. It will probably be the same for AMD’s Zen/Vega Zen/Navi based HPC/workstation/server APUs on an interposer and AMD/SK Hynix HBM2 memory! Looks like GDDR5 will have to do for all but the flagship SKUs, unless the HBM2 production can be ramped at a faster pace!
HBM2 will follow to where that markups/margins are the highest!
Money talks, CPU only MOOKs walk! GPUs rule! RUN AWAY you Knights of Phi, your Xeon’s defeat is really nigh! Both the Red and Green Armies are on the Async Move,
and your CPU Kludge has lost its groove!
“Tesla Pushes Nvidia Deeper Into The Datacenter”
http://www.nextplatform.com/2016/05/16/tesla-pushes-nvidia-deeper-datacenter/
“FinFET Scaling Reaches
“FinFET Scaling Reaches Thermal Limit”
“Advancing to the next process nodes will not produce the same performance improvements as in the past.”
http://semiengineering.com/dennards-law-and-the-finfet/