Grigorii Entin

Controlling the seed is probably also relevant in context of property-based testing, like in [SwiftCheck](https://github.com/typelift/SwiftCheck).

I face this issue as well on M1 running BigSur 11.3.

> We are having the same problem. For us it works too in 2.207.0, the breaking change is definitely in 2.208.0. Yeah, I can confirm it's a regression in 2.208.0....

@halonsoluis Out of curiosity, why can't we have Resolvers that are plain values? I wonder if there is really a "retain cycle problem" with value types? I.e. I can imagine...

> In my case snapshots began rendering identically under Apple Silicon after I added 'Host Application' for unit test targets. I'm not 100% sure, but in our case we get...

I wonder if any kind of solution that employs imprecise matching would result in loads of "modified" snapshots generated every time they're re-recorded on a "mismatching" platform: as far as...

@ArielDemarco >We tried to use less precision but some of the snapshots failed (even with 0.9 😓 ) I can only recommend taking a look at this, that's a fuzzy...

@agurtovoy Well, even though we've been using it for quite some time in "production", I would like to get a confirmation from somebody else, that it's useful and works for...

@agurtovoy Re: SNAPSHOT_RECORDING - I also thought about a separate PR, but now I think that it's related to another feature "re-implemented" here: being able to *delete* the obsolete snapshots...

As far as I understand README, libetpan is stored in binary form here. Given that M1 was not a thing 2 years ago, and that this project has not updated...