When will QCA release AC3100 chipset?

[Nerva]

From the reviews I've been reading on SNB, the Netgear R7800 and QCA9984 chipset are clearly superior to the current Broadcom-based products -- but the one thing it doesn't currently do is 1024-QAM. I thought I read somewhere that QCA planned to support 1024-QAM for 802.11ac, so I am curious if they have another chipset coming soon that will add that feature?

 

[sfx2000]

From the reviews I've been reading on SNB, the Netgear R7800 and QCA9984 chipset are clearly superior to the current Broadcom-based products -- but the one thing it doesn't currently do is 1024-QAM. I thought I read somewhere that QCA planned to support 1024-QAM for 802.11ac, so I am curious if they have another chipset coming soon that will add that feature?

Keep in mind that 1024QAM is not in the 802.11ac specification, and is a proprietary implementation by Broadcom...

Atheros, back in the day, used to push the specs pretty hard (anyone remember SuperG?), but Qualcomm tends to be a bit more strict on what they push out to the market.

In any event, most folks will never see QAM1024 in the real world, conditions have to be perfect, and while this can be done in a lab, it's hard to achieve that in the real world.

 

[Nerva]

Good to know.

By "proprietary implementation by Broadcom", do you mean Broadcom's AC3100 routers will only use 1024-QAM with Broadcom clients, even if QCA adds their own support for it?

 

[sfx2000]

Good to know.

By "proprietary implementation by Broadcom", do you mean Broadcom's AC3100 routers will only use 1024-QAM with Broadcom clients, even if QCA adds their own support for it?

Depends on what Broadcom wants to do with it - it's outside of the spec, folks can either license it from there, or reverse engineer...

 

[thiggins]

1024 QAM is another example of Broadcom attempting to throw FUD into the consumer Wi-Fi market. I've yet to see a device reach the advertised maximum link rate and it's unlikely I will.

1024 QAM requires devices in the RF chain like switches and power amplifiers to have extremely low error-vector-magnitude (EVM), lower than devices in current use today have.

Even if 1024 QAM works, it requires very strong signals to do so. So any throughput gain would be seen in the same room or MAYBE next room use.

QCA followed Broadcom in implementing 256 QAM, which Broadcom used to create AC1900 vs. AC1750. Given QCA's lead in producing WORKING MU-MIMO, I don't know why they would play the game again.

I hope the focus on multi-AP systems like eero, Luma, AmpliFi and others soon to come gets people off the bigger number is better train. It really doesn't solve anything besides keeping new revenue coming in for router makers.