This is the group with the most thankless job, the one which is directly customer facing. Sadly the one which is also most vulnerable if there is a slowdown...
10 replies (most recent on top)
Anonymous115521, I don't think the SS business will go to zero. It wont be close to 100% but definitely not zero either. What Samsung will do, provided the 820 is really better, is split the release of future product launch (IE S7) into one variation that has their own app processor and one that variation that has the 820. Until the S6, they did that on the S5, where the ones sent over here was had Q's chip and ones in the local asia markets had their own chip in it. It really depends on how much better the 820 is versus their own chip...As far as LTE modem, they have their own baseband too, and that is probably a bigger issue. Because if it's doing alright, there's really no compelling reason for them to buy anyone else's....The bigger problem, however, is Samsung's declining market share. The S6 sales wasnt as good as expected (partly due to supply issues). But apple is just killing Samsung. And you have a bunch of chinese companies like Xiaomi, ZTE, Huawei that aren't that much worse than Samsung, and costs really cheap.
Anonymous115429, Easier said than done....Once you repeatedly pile on top over and over again, it's really hard to start from a clean slate..And because when a product is in this state, management never will want to invest time and resources into making something "correct" when there are pending potential customers that want everything now.
It's actually funny seeing this at a customer engagement. You have Qualcomm bring a fleet of 15-20 people...And you have mediatek bring maybe 5-6 people. Multiply that by the number of customers, and you can start to understand why other companies are able to run at a much lower cost. I mean, just consider the travel cost alone for 15-20 people, versus MediaTek already based in Asia bringing in 5-6 people. Not to mention that those 15-20 people end up flying business class and stay at a 4-5 star hotel, for a pretty long time. MediaTek on the other hand always flies it's employees coach... It's the little things, but you can tell just from something as simple as this how drastically different operation costs must be.
So instead of having so many people in CE, why not use people to creatte and continually update the refference design? Or would that make too much sense?
CE will be affected as the projects with SS are reduced to close to 0
CE will be affected as the projects with SS are reduced to close to 0
Actually, with Qualcomm, you do need a large CE group. The reason being is that at least when it came to platform integration on the mobile phone, it's a total pain in the ass. No seriously....That's one of the reasons why MediaTek can win in many cases. They have a pretty good reference design, and not nearly as much supported is needed. In some cases, customers are willing to pay more for their reference design.
No customers == No CE
CE already went through a round of layoffs last time around. An acquaintance of mine got laid off and found another job within qcom (he got severence pay and a signing bonus too... wtf).
tbqh, I don't see the point of a huge CE team. I'll grant you that there are few good CEs here and there, but for the most part they just collect logs from customers and give to devs. I don't remember the last time that they even triaged an issue, instead of just dumping the logs in devs head.
(no offense to any CEs reading this but I haven't had the best experience with your team)
Engineers need CE to pass the bucket to get enough work to do.
CE would be fine.