Record to internal drive?

[dominicperry]

Originally Posted by Ben Jenssen

Pedantic is the word.

I was being polite. But I will be precise instead. You were incorrent, which is misleading when people are trying to work things out.

[Ben Jenssen]

Originally Posted by dominicperry

I was being polite. But I will be precise instead. You were incorrent, which is misleading when people are trying to work things out.

Oh, come on, please!

[Oliver M]

Here is my point of view:


Since 20+ years I record everything to external drives, SSDs since quite a while, keeping the internal drive(s) as clean as possible. It is much easier to just backup this way and it is also less troublesome. My 2 cents.

[JFreak]

Originally Posted by Oliver M

Here is my point of view:


Since 20+ years I record everything to external drives, SSDs since quite a while, keeping the internal drive(s) as clean as possible. It is much easier to just backup this way and it is also less troublesome. My 2 cents.

Yes. I have just been lazy buying that external SSD that has 3.5GB/sec throughput


Sent from my iPhone using Tapatalk

[gthm]

Hmm. Recording to the internal drive. Kind of makes sense with the speeds we enjoy today. Still weird a concept though hehe.. I am old.

What about VI content, samples etc.? Don't tell me those on the internal as well?

Anyone?

[Darryl Ramm]

Originally Posted by gthm

What about VI content, samples etc.? Don't tell me those on the internal as well?

This has been discussed on DUC many hundreds of times already.

Sure almost every Pro Tools user should be able to put *everything* on these superfast internal PCIe/NVMe SSDs. Sessions, samples, video, etc. Limited only by space/how much they are willing to pay Apple tax for those expensive internal SSDs. The performance of these modern PCIe/NVMe SSSs bears no relation to anything in the past that required separate drives. And largely the best way to slow down the IO is to put stuff on external drives, including external SATA based SSDs... only external Thunderbolt 3 SSDs with internal NVMe/PCIe SSDs are in the same ballpark as the internal NVMe/PCIe SSD performance. ANd at least for sessions, disk cache largely masks lots of disk performance, even if the NVMe/PCIe SSD was not blindingly fast to start with... but you typically don't *need* to have disk cache enabled, but it's' a good thing (set to a size to cache the session not left at "normal").

[EGS]

Originally Posted by Darryl Ramm

...disk cache largely masks lots of disk performance...

So why does the read/write speed of a drive even matter at all? You can run a session from a flashdrive for that matter.

[Darryl Ramm]

Originally Posted by EGS

So why does the read/write speed of a drive even matter at all? You can run a session from a flashdrive for that matter.

Disk cache can help a lot, but does not help with VI samples or streaming video and it can slow session startup times and in cases where the cache does need to flush stuff out as you are working if the drive is slow you can still see glitches. And things like session startup, and maybe boot times matter. Backup times as well, but often that is going to HDD or NAS… (and off topic I do *not* recommend SSD for archival storage).

Many USB thumb drives are very different than a high performance PCIe/NVMe drive, especially when running at heavy sustained load/high IO queue depth/thermal load… and can have very much shorter wear lives. But sure you can record to them, for light workloads and good thumb drives you don’t even need disk cache. But, but, but… I would generally not do that. I note all the care that RME takes with qualifying USB thumb drives and larger USB SATA drives for my RME Fireface UFX+ And they give users there great tutorials about not all thumb drives are equal. (That system does not have equivalent to disk cache, but also is not running samples or video off the drive… but it does highlight the huge variance in performance of USB thumb drives). And we see do see similar issues with “SSD” drives in general where their performance can vary by huge factors, where some drives are cheap and nasty, poorly designed/implemented and often dishonestly marketed but they are “SSD” so unlucky consumers think they will be fast/reliable/have long life etc. Probably the best thing that has helped SSD adoption is Apple’s use of quality well implemented PCIe/NVMe in their systems. And that has helped show what is possible and seed widespread adoption.

One trap with recording Pro Tools to a thumb drives I see people falling into is not reformatting them to a supported file system, folks are risking problems recording to the default exFAT file system. But sure stuff there might work for a while…

Quality thumb drives are magical tools for quickly moving stuff between computers, carrying tools or installers with you etc. I love the high-end SanDisk Extreme Pro ones, but also note if you want compact portable SSD storage the larger but still small Samsung T7 drives are impressive for their size/cost, they are not Thunderbolt, but are NVMe over USB and do have good thermal bonding to their aluminum case (but no heat fins)… seems a lot better thermal design than the Samsung Thunderbolt X5 portable drives. A T7 fits in your pocket, but it’s easy to lose the USB cable so not quite a thumb drive in carry with you always convenience. And a T7 makes a nice session drive on just about any computer with USB 3.

[EGS]

Originally Posted by Darryl Ramm

Disk cache can help a lot, but does not help with VI samples or streaming video ... (and off topic I do *not* recommend SSD for archival storage)...

+1. I set disk cache to 8GB and it works. For VI samples or video in the session, use a good internal SSD. For archives, I'm using spinners.

[StudioTalk]

The days of needing separate drives is over and has been for a while now. As a matter of practice and workflow, I keep mine on external NMVe Drives for portability, and to ensure my internal drive doesn't fill up.