By: David Kanter (dkanter.delete@this.realworldtech.com), July 14, 2020 8:13 am
Room: Moderated Discussions
Doug S (foo.delete@this.bar.bar) on July 13, 2020 11:25 pm wrote:
> Maynard Handley (name99.delete@this.name99.org) on July 13, 2020 10:37 pm wrote:
> > I used to think this (mask costs) was a big deal. I’m no longer convinced.
> > A 5nm mask set supposedly costs $15M. For comparison Apple sells ~20M Macs per year.
>
>
> Do you have a source for that?
Maynard's numbers sound about right. But remember that masks don't last forever, and you typically need respins.
> I was under the impression a 7nm mask set was in the ballpark of
> $30 to $60 million, though since TSMC's 5nm is using EUV there are fewer masks required (due to
> the lack of double/quad patterning steps) but I'd be surprised if it was half as many since the
> DP/QP stuff is only done for a few layers at the finest pitch. Not that I dispute your overall point
> that mask costs aren't a big deal when you have Apple's volumes and the Mac's cost structure.
People are still using double patterning. I suspect people will still continue to use quad patterning in some contexts.
> What I wonder about though - and maybe those in the industry can enlighten me - is that you don't really
> just need ONE mask set, right? No design is perfect out of the gate, you do the original design, prepare
> a mask set, run a hot lot and get samples back and test them and you find problems. Then you update
> the design to fix problems and you need new masks. You won't change every layer so the entire mask
> set doesn't need to be replaced, but maybe 1/3 or 1/2 of it does since the layers at the finest pitch
> will be the ones most likely to change. And you may do this through several revs.
I'd expect most folks need 2 mask sets. Intel tends to iterate more, but they also have their own mask sets that are made internally.
> The other thing I'm unclear on is if you want to produce something at high volume using multiple
> lines at the fab, do you need separate mask sets for each line?
We are talking about low volume stuff, where you wouldn't need more than a single line.
David
> Maynard Handley (name99.delete@this.name99.org) on July 13, 2020 10:37 pm wrote:
> > I used to think this (mask costs) was a big deal. I’m no longer convinced.
> > A 5nm mask set supposedly costs $15M. For comparison Apple sells ~20M Macs per year.
>
>
> Do you have a source for that?
Maynard's numbers sound about right. But remember that masks don't last forever, and you typically need respins.
> I was under the impression a 7nm mask set was in the ballpark of
> $30 to $60 million, though since TSMC's 5nm is using EUV there are fewer masks required (due to
> the lack of double/quad patterning steps) but I'd be surprised if it was half as many since the
> DP/QP stuff is only done for a few layers at the finest pitch. Not that I dispute your overall point
> that mask costs aren't a big deal when you have Apple's volumes and the Mac's cost structure.
People are still using double patterning. I suspect people will still continue to use quad patterning in some contexts.
> What I wonder about though - and maybe those in the industry can enlighten me - is that you don't really
> just need ONE mask set, right? No design is perfect out of the gate, you do the original design, prepare
> a mask set, run a hot lot and get samples back and test them and you find problems. Then you update
> the design to fix problems and you need new masks. You won't change every layer so the entire mask
> set doesn't need to be replaced, but maybe 1/3 or 1/2 of it does since the layers at the finest pitch
> will be the ones most likely to change. And you may do this through several revs.
I'd expect most folks need 2 mask sets. Intel tends to iterate more, but they also have their own mask sets that are made internally.
> The other thing I'm unclear on is if you want to produce something at high volume using multiple
> lines at the fab, do you need separate mask sets for each line?
We are talking about low volume stuff, where you wouldn't need more than a single line.
David
| Topic | Posted By | Date |
|---|---|---|
| Skylake-SP area breakdown | David Kanter | 2020/07/12 06:13 PM |
| Skylake-SP area breakdown | anon2 | 2020/07/12 07:01 PM |
| Skylake-SP area breakdown | Travis Downs | 2020/07/12 08:02 PM |
| Skylake-SP area breakdown | anon | 2020/07/12 08:44 PM |
| Skylake-SP area breakdown | Travis Downs | 2020/07/12 08:03 PM |
| Skylake-SP area breakdown | David Kanter | 2020/07/12 08:20 PM |
| To elaborate | David Kanter | 2020/07/12 08:22 PM |
| To elaborate | Travis Downs | 2020/07/13 07:03 AM |
| To elaborate | Anon | 2020/07/13 07:36 AM |
| To elaborate | Adrian | 2020/07/13 01:45 PM |
| To elaborate | Anon | 2020/07/13 02:06 PM |
| To elaborate | Chester | 2020/07/13 08:30 PM |
| Alternatives Implementations | Kyle Siefring | 2020/07/13 06:02 PM |
| Alternatives Implementations | Travis Downs | 2020/07/13 08:41 PM |
| Alternatives Implementations | Maynard Handley | 2020/07/13 10:37 PM |
| Alternatives Implementations | Doug S | 2020/07/13 11:25 PM |
| Mask costs | David Kanter | 2020/07/14 08:13 AM |
| Alternatives Implementations | tarlinian | 2020/07/14 08:22 AM |
| Alternatives Implementations | Doug S | 2020/07/14 10:03 AM |
| Alternatives Implementations | Maynard Handley | 2020/07/14 10:12 AM |
| Alternatives Implementations | Maynard Handley | 2020/07/14 10:10 AM |
| Alternatives Implementations | Doug S | 2020/07/14 10:47 AM |
| Alternatives Implementations | Brett | 2020/07/14 01:38 PM |
| Alternatives Implementations | tarlinian | 2020/07/14 02:30 PM |
