Hacker News new | past | comments | ask | show | jobs | submit login

There's not really an alternative.

Broadcom is one of the few options out there for ARM SoC. Rpi could dump a bunch of money into making their own SoC, but that would really ballon the costs.




They couldn't use Rockchip or Allwinner or Texas Instruments chips? Sure there is some work they would have to do all over again getting the drivers working and reliable, but in the long run the partnership could be far better for the company.


They should partner with qualcomm, not broadcom.

Too bad it's because of the former broadcom employees within the foundation.


Qualcomm is the single worst company to work with. They try to negotiate % of revenue deals instead of cost-per-unit deals and they make Broadcom look like Stallman-level open source fanatics.


> They should partner with qualcomm, not broadcom.

lol, are you unaware of how qualcomm treats all of it's customers, including megacorps?


I'm guessing you're not a fan of Google Wear watches or Microsoft Surface ARM at all or you'd be more aware of how Qualcomm treats their gift horses worse than I've seen any company. We are talking 10 year old mediocre designs being shown off as their flagship parts in some cases here and I'd expect if they did turn over the Pi to Qualcomm you'd get the best 2015 $100 phones had to offer as the SoC


They should produce boards from multiple vendors' SoCs.


This doesn’t work in practice since your costs scale with vendor count and theirs don’t.

In reality being a single big customer for one vendor is far more influential than being a small one with other options.


In reality being a single big customer for one vendor is far more influential than being a small one with other options.

In what world? the one where firms don't compete?

If other firms are explicitly on the table for the currently negotiated and next generation RPi, Broadcom will have more incentive to meet the RPi Foundation's specific needs. If Broadcom knows that RPi is not talking to other vendors, then their negotiating position is stronger.


In the real world. I’ve been working for large vendors for literal decades and I can tell you outright that you own us if you’re a big customer, even if we are the only vendor in sight, and no one gives a shit if you’re spreading out your business.

This is 100% a natural consequence of the need to make your quarter. A bigger deal is much more influential than a small one, consistently. Sales incentives also do this.

People who have not actually worked in the industry with a solid eye for how this works often think that having multiple vendors is a good answer, but it’s not. Having optionality helps, but in the end it’s all about deal size.


You generally get bulk discounts. The more vendors you have, the more expensive each is.


Also think of the added development expenses, which might even overshadow forgone bulk discounts


For different SoCs? You would just choose one vendor per device generation (but always evaluating multiple). It is a new order anyways.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: