[802.3_ISAAC] AW: Presentation for the July Plenary

["Matheus Kirsten, EE-352" <Kirsten.Matheus@xxxxxx>]

Hello Max,

 

thank you for sharing the slides with the group. Of some of the aspects you are showing, I have a fundamentally different understanding and therefore cannot follow your arguments or conclusion.  

 

It looks to me that you are neglecting the difference between link and line rate and the levels in the processing chain at which the comparisons happen (i.e. at PHY or MAC level).

 

On slide 6, e.g. I do not understand why you set the TDD cycle to 125.3us. If you do that, you would not be able to achieve a 10G/100M (average) link rate any more. To ensure the same throughput, a TDD system would always have a somewhat higher line rate (compared with a system with the same processing overhead, more in my latency presentation) and an FDD system would always occupy a higher frequency range. The link rate and what the MAC sees, is unaffected.

 

I do not understand, why you tie, what happens on PHY level to what happens at the MAC level. That seems unnecessary to me.

 

I see two ways of achieving asymmetric link rates (also somewhat addressed in the buffering presentation) between MAC and PHY ( I remember a presentation from William Lo also discussing those).

 

1. By having an XGMII in one direction and an MII in the other. This is full duplex at MAC level and may be used by an FDD or a TDD system. The MAC would not (need to) know! It would simply forward the data at 100Mbps. Whether the PHY holds the data and then sends it faster in a bulk or not, the MAC should not care.
2. By having an XGMII in both direction and an LPI(-like) client/higher layer ensures that the average link rate in the US direction is also just 100M. This would be the case for an FDX/EEE system or it may also be adopted for a TDD approach. Again, the MAC would not know the difference.

 

So, for what reason would the MAC need to know the PHY timing?

 

In general: I see the solution, we are discussing at dm, for the last hop to the camera. Also, I do not see the TAS in cars, independent from the PHY. Maybe some of the different understandings come from there?

 

Kind regards,

 

Kirsten

 

Von: Max Turner <max.turner@xxxxxxxx>
Gesendet: Dienstag, 2. Juli 2024 09:32
An: STDS-802-3-ISAAC@xxxxxxxxxxxxxxxxx
Cc: Jon.Lewis@xxxxxxxx; Matheus Kirsten, EE-352 <Kirsten.Matheus@xxxxxx>; hossein.sedarat@xxxxxxxxxxxxxx
Betreff: Presentation for the July Plenary

 

Sent from outside the BMW organization - be CAUTIOUS, particularly with links and attachments.

Absender außerhalb der BMW Organisation - Bitte VORSICHT beim Öffnen von Links und Anhängen.

---

Dear all!

 

Please find attached my slides for the .3dm session during the July Plenary.

 

Feedback and questions up front are very welcome!

 

Best

 

Max

 

 

--

Max Turner, Dipl.Phys.

Automotive Network Architect

Ethernovia BV

Utrechtseweg 75

3702AA Zeist

The Netherlands

https://www.ethernovia.com/

c-de: +49 177 863 7804

c-nl: +31 685 386 449

---

To unsubscribe from the STDS-802-3-ISAAC list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-ISAAC&A=1