I'd estimate a RX560 or just below a GTX 1050 since that's what the revamped Vega CUs in Renoir performed close to, and Cezanne is just reusing Vega. Clock speeds are up a little though.any idea of gpu equiv ?
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
I'd estimate a RX560 or just below a GTX 1050 since that's what the revamped Vega CUs in Renoir performed close to, and Cezanne is just reusing Vega. Clock speeds are up a little though.any idea of gpu equiv ?
I'd estimate a RX560 or just below a GTX 1050 since that's what the revamped Vega CUs in Renoir performed close to, and Cezanne is just reusing Vega. Clock speeds are up a little though.
I suspect that the integrated (Vega based) GPU chip may be so cheap to produce ($10-$20?), that it just isn't worth making a version without it
An RX560 is faster because the IGP is bottlenecked by memory bandwidth. This is a comparison of the Ryzen 7 4750G against the RX560:I'd estimate a RX560 or just below a GTX 1050 since that's what the revamped Vega CUs in Renoir performed close to, and Cezanne is just reusing Vega. Clock speeds are up a little though.
AM4 is only dual channel, unless you mean that the memory controller might be able to do some interleaving with the extra DIMMs?Curious how the 5700G iGPU with 4 ram sticks would perform. (More bandwidth).
AM4 is only dual channel, unless you mean that the memory controller might be able to do some interleaving with the extra DIMMs?
Does the 4700g have the same iGPU as the 5700g?
The 4750G and 5700G share a very similar GPU, the difference is just a small frequency change, the massive benefit of the 5700G is the 16MG L3 cache which makes a significant difference in performance when you are memory bandwidth starved.
Have AMD given any technical reasons why these chips are monolithic instead of adding the GPU portion as a chiplet?
For the chiplet design, going that route for expensive processors actually helps with costs, yields, and frequency binning. This has also enabled AMD to launch these products earlier, with the end result being better per-core performance (through binning), taking advantage of different process nodes, and providing an overall chip with more die area than a single chip can provide.
The downside of this chiplet design is often internal connectivity. In a chiplet design you have to go ‘off-chip’ to get to anywhere else, which incurs a power and a latency deficit. Part of what AMD did for the chiplet designs is to minimize that, with AMD’s Infinity Fabric connecting all the parts together, with the goal of the IF to offer a low energy per bit transfer and still be quite fast. In order to get this to work on these processors, AMD had to rigidly link the internal fabric frequency to the memory frequency.
With a monolithic design, AMD doesn’t need to apply such rigid standards to maintain performance. In Ryzen Mobile 4000, the Infinity Fabric remains on the silicon, and can slow down / ramp up as needed, boosting performance, decreasing latency, or saving power. The other side is that the silicon itself is bigger, which might be worse for frequency binning or yield, and so AMD took extra steps to help keep the die size small. AMD was keen to point out in its Tech Day for Ryzen Mobile that it did a lot of work in Physical Design, as well as collaborating with TSMC who actually manufactures the designs, in order to find a good balance between die size, frequency, and efficiency.
Because these are still essentially mobile APUs and chiplets don't work well in mobile right now.Have AMD given any technical reasons why these chips are monolithic instead of adding the GPU portion as a chiplet?
That would make sense. So effectively the upcoming 5600g and 5700g are mobile chips that didn't make the grade.Because these are still essentially mobile APUs and chiplets don't work well in mobile right now.
Likely. I'm not sure if there's any difference between a 5700G and a 5980HX besides the binning; if you can't get 3.3GHz base, 4.8GHz boost and 2.1GHz GPU out of the 8 core 8 CU package at 45W, let it run 3.6 GHz base at 65W and slap it into a AM4 package as a 5700G.That would make sense. So effectively the upcoming 5600g and 5700g are mobile chips that didn't make the grade.
That would make sense. So effectively the upcoming 5600g and 5700g are mobile chips that didn't make the grade.
Fixed that for you.That would make sense. So effectively the upcoming 5600g and 5700g are mobile chips that didn't make the grade for mobile.
On the AT threads on the 3D stacked cache thread:Have AMD given any technical reasons why these chips are monolithic instead of adding the GPU portion as a chiplet?
On the AT threads on the 3D stacked cache thread:
https://forums.anandtech.com/threads/64mb-v-cache-on-59xx-zen3-average-15-in-games.2594096/
And the Zen4 speculation thread:
https://forums.anandtech.com/threads/speculation-zen-4-epyc-4-genoa-ryzen-6000.2571425/
there is some talk about the power requirements of chiplets with a PCB like the current AM4 desktop parts and the 3D stacked cache.
The distances and power required to talk to an off-chip part over Infinity Fabric is huge compared to talking to the 3D stached cache, the distances are higher, the voltages too and there, AFAIK, the need to translate everything from parallel to serial.