PC Perspective Podcast #343 – 04/02/2015
Join us this week as we discuss DX12 Performance, Dissecting G-SYNC and FreeSync, Intel 3D NAND 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
- RSS – Subscribe through your regular RSS reader
- MP3 – Direct download link to the MP3 file
Hosts:Ryan Shrout, Jeremy Hellstrom, Josh Walrath, and Allyn Malventano
Program length: 1:27:10
-
Week in Review:
-
News item of interest:
-
Hardware/Software Picks of the Week:
-
Jeremy: Just pretend it is Half Life 3
-
Closing/outro
Subscribe to the PC Perspective YouTube Channel for more videos, reviews and podcasts!!
Do you have an AMD FX-8350
Do you have an AMD FX-8350 that you could run the 3D Mark API overhead test on? DX12 is supposed to reduce cpu overhead, so I am curious as to how an AMD cpu will do. They actually perform quite well in well multi-threaded applications. Can you disable cores on an FX-8350 such that only one core from each module will be active?
With the free sync vs. gsync
With the free sync vs. gsync comparison, it seems like free sync has serious issues below the monitors variable refresh window. Also, the display controller may not be accurately correcting overdrive (increasing the voltage used to make liquid crystals change state to increase response time), for variable refresh rates. It seems that the gsync module may be doing this, but for free sync this would be left up to the display controller to accurately modify overdrive settings depending on the current refresh rate. This can probably be fixed in the display controller. For fixing the issues below the variable refresh window, AMD may be able to do this in the driver, but it might not be easy without having specialized hardware.
Josh, calculus vs. …?
Josh, calculus vs. …? Infinite slices…? It sounds like you are trying to describe analog vs. digital (or at least continuously variable vs. quantized). An LCD actually does move from one color to next somewhat smoothly; it is not instantaneous. Moving smoothly from one color to the next isn’t useful though. Eyes are not even close to unlimited refresh rate. You can move your hand fast enough to turn it into a blur as far as human vision is concerned. You do not need to go to a much higher refresh rate before it would be outside the abilities of human vision to detect. There are other technology which can produce refresh rates which are outside humans ability to detect. DLP projectors actually produce colors by strobing the primary colors sequentially; the color is displayed by varying the duty cycle. This takes advantage of the persistence of human vision. It is somewhat annoying that DLP projection televisions died due to not being flat enough. The image quality on the last produced models of projection TVs was significantly higher than common LCDs. Most TN panels have terrible color accuracy. OLEDs also solve this problem since they can be strobed much faster than a human can detect but there has been a lot of talk with few actual products. The quantum dot display tech may make high refresh and good color not mutually exclusive like we currently have with TN vs. IPS LCD displays. Also, we may actually get OLED displays soon (hopefully).
Nope, not analog. Though the
Nope, not analog. Though the idea is similar, but we are still dealing with frames which in ways still is essentially digital in nature. If the source was analog, then yes, we would be talking analog. But the source compiles frames, so my "infinite slices" is an equation that links the two digital frames with a smooth conversion of color on that particular pixel. Though I have to ask… is what I am saying making sense?
Arthur
Arthur Terrabyte
DroolTekk.com
mr. Josh.. how come mobile
mr. Josh.. how come mobile GPUs can use lower processes for their GPU? can’t we just cram more stream processors in and underclock the gpu by a lot??? ALSO, the next bottleneck to CPU is memory bandwidth with a 4096bit bus i’m thinking AMD is ahead by much?? yes?? no??
*mobile SOCs i mean
*mobile SOCs i mean
Hello PCPER,
We wanted to say
Hello PCPER,
We wanted to say thank you very much for using our SIRS-E 3 Channel RGB LED DMX Controller R2 at your studio and showing it on your Live Podcast. If possible could Mr. Shrout or someone on your staff contact us via email or call us at your earliest convenience. We would like to discuss your experience with the controller and ask you a couple of questions if possible.
Regards,
SIRS-E