Question AMD Phoenix/Zen 4 APU Speculation and Discussion

Page 83 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
LOL. Whatever helps you sleep better at night smh.

I gave technical explanations, ask any capable engineer and he ll tell you that i explained the things correctly in my previous post.

Actually you didnt even realise that the two triangles in Semiwiki schematic are symbolizing fet transistors, let alone which terminals (out of three for each fet) are supposed to be used and in which way, so much for your enginering credentials.
 

Hitman928

Diamond Member
Apr 15, 2012
6,695
12,370
136
I gave technical explanations, ask any capable engineer and he ll tell you that i explained the things correctly in my previous post.

Actually you didnt even realise that the two triangles in Semiwiki schematic are symbolizing fet transistors, let alone which terminals (out of three for each fet) are supposed to be used and in which way, so much for your enginering credentials.

Straw man after straw man. But I'll tell you what, if you can read the following paper and let me know where they fail in accounting for the Miller cap, I will gladly have an actual technical discussion with you.

 

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
Straw man after straw man. But I'll tell you what, if you can read the following paper and let me know where they fail in accounting for the Miller cap, I will gladly have an actual technical discussion with you.


I ve not registration in this site, beside the straw is to elude the debate by presenting publicly unavailable docs that you dont forcibly understand.

Stick to wiki article and tell me what was wrong in my explanation if you can, but you know that you cant, hence you re trying to dragg the discussion in other issues of ED,
that s just childish behaviour out of desperation of being caught pants down.


What is wrong in this post, tell us...
 

Khato

Golden Member
Jul 15, 2001
1,279
361
136
I ve not registration in this site, beside the straw is to elude the debate by presenting publicly unavailable docs that you dont forcibly understand.
Surprising, all professionals in the field I know have IEEE access through their employer.
 
  • Like
Reactions: Hitman928

Hitman928

Diamond Member
Apr 15, 2012
6,695
12,370
136
I ve not registration in this site, beside the straw is to elude the debate by presenting publicly unavailable docs that you dont forcibly understand.

Stick to wiki article and tell me what was wrong in my explanation if you can, but you know that you cant, hence you re trying to dragg the discussion in other issues of ED,
that s just childish behaviour out of desperation of being caught pants down.

I assumed, since you say you've designed hundreds (or was it thousands?) of circuits that you would have access to circuit design papers (like every other circuit designer I know). I can't comment on your blog post link because the actual white paper link inside the blog post goes no where, so there's not really anything to comment on outside of the obvious.

You can refer back to our last discussion for my problems with your comments on digital design, like I said, I'm not going through that again. Have a nice weekend.
 
  • Like
Reactions: Markfw

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
I assumed, since you say you've designed hundreds (or was it thousands?) of circuits that you would have access to circuit design papers (like every other circuit designer I know). I can't comment on your blog post link because the actual white paper link inside the blog post goes no where, so there's not really anything to comment on outside of the obvious.

You can refer back to our last discussion for my problems with your comments on digital design, like I said, I'm not going through that again. Have a nice weekend.
in another site i made a basic math demonstration about some limiting factor in opamps linearity, and engineers with 30 years of experience were unaware of this limitation, and that s people working not only in reviews but also in firms like Analog Devices, not to name the famed engineer who was in the dicussion, and yet, i can only aknowledge that they have much more knowledge than me, so no one can pretend that he knows everything.

Surprising, all professionals in the field I know have IEEE access through their employer.

Who told you that i ve still any activity in this domain.?.


But that s another debate, i m still waiting for proves that i was wrong here :

 
Last edited:

Asterox

Golden Member
May 15, 2012
1,058
1,864
136
As expected, especially when comparing R5 5600G vs R5 8500G with a combination of 2x Zen4 and 4x Zen4c CPU cores.

2024-01-28_162901.jpg
 

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
As expected, especially when comparing R5 5600G vs R5 8500G with a combination of 2x Zen4 and 4x Zen4c CPU cores.

View attachment 92624

Even if perfs are substancialy better than the 5000 series price is instrumental for sales success.

The last two week or so Mindfactory did dramatic prices cuts on the 5600G/5700X/5800X wich did hit historical lows, result is that with 20-25% price cuts sales increased by 50-100% for these SKUs.

They also did such a price cut on the old 5950X wich was on the bottom of the sales list and its sale increase by 2-3x matching the 7950X3D wich was a good seller.

This was also the case with the 7600/7600X/7700x wich were also sold cheaper for a week and witnessed quite higher sales, so it seems that elasticity of the demand/price is quite high in the 100-300€ range, guess that AMD s expected prices for these new SKUs should be lower than what was first announced if they want to have decent sales.

 

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
That s about 40% of PC sales for DT CPUs, but this number is not accurate since it s sales of big names like HP or Lenovo, this doesnt include DIY PCs where there s no laptops, so DT CPUs amount surely to a good 50% of all CPUs sold.
 

Tigerick

Senior member
Apr 1, 2022
846
799
106
That s about 40% of PC sales for DT CPUs, but this number is not accurate since it s sales of big names like HP or Lenovo, this doesnt include DIY PCs where there s no laptops, so DT CPUs amount surely to a good 50% of all CPUs sold.
No, all these years the split is about 80:20. Desktop sales are amounted to 20% of total volume. And the majority goes to OEMs, DIY market is pretty small; it is in single digit of percentage...
 
Jul 27, 2020
28,019
19,126
146
I would love to know the German thought process that's leading to those Intel CPU sales. Are these illiterate Germans or is there a valid reason for them to go Intel?
 
Jul 27, 2020
28,019
19,126
146
Why is AMD being shy with the slower APUs using the 4c dies? Why no review of that? I summarily object to your shyness, AMD!
 

FlameTail

Diamond Member
Dec 15, 2021
4,384
2,761
106
Right that the numbers shrinked, but that s 28.5%, 81.6M for DT and 286.2M globally, that s for the sales of big names, so not including DIY wich is widespread in countries with lower incomes than western countries.
I really doubt there are a lot of people doing DIY builds.
 
Jul 27, 2020
28,019
19,126
146
I really doubt there are a lot of people doing DIY builds.

1706542864117.png
DIY is just cheaper in this region. This mostly includes shops doing the builds for their customers.
 

Tigerick

Senior member
Apr 1, 2022
846
799
106

View attachment 92692
DIY is just cheaper in this region. This mostly includes shops doing the builds for their customers.
You shouldn't check the sales figures as we all know the MB pricing has been doubled since COVID.


Here is the MB shipment from top four OEMs, namely ASUS, Gigabyte, MSI and Asrock. They represent almost top brand MB manufacturers in DIY market including China. The total volume shipments have dropped from 44.5 millions (2021) to 31.3 millions (2022). 31.3 millions out of 286.2 millions total PC means about 11% of PC market, sound about right...

PS: I am looking for 2023's shipments, will update here if I find it
 
Last edited:
  • Like
Reactions: Tlh97 and Hitman928

Asterox

Golden Member
May 15, 2012
1,058
1,864
136
8500g is low-end. That's why.
No it is not, currently it is the cheapest Zen 4 6/12 APU for AM5 socket.
Even in an AM4 socket, today you can go much lower than the R5 5600G.

If R5 8500G is low-end, where does this Intel socket 1700 2/2 CPU belong in which class? :mask:


4/8 Ryzen 3 3100 new or used for 50-60$ vs new 2/2 Celeron G6900 for 67$ or euro!That's low-end, but the Celeron actually fell even lower in the hole.