Another Laptop Spec Question

Not exactly a Laptop, as I rather need to replace my NUC PC by something with a similar form factor and more RAM, but I found, this which could possibly do the job:

I didn’t see a DPC Latency for this model, but it performed well with the older IT 8.

1 Like

It seems like some of this is a crap shoot.

I think I got lucky my Dell XPS 13 that I converted to this use. (Or maybe it is more accurate to say I was not among the percentage of GP users who are “unlucky”).

Fingers cross with this new Lenova Thinkpad 16…

1 Like

I’m interested in hearing are you satisfied with the results :slight_smile:

1 Like

You know you’ll hear from me, like it or not! Hah!



PS: I’m still spending stupid time researching this (event though I already bought it).

I “think” this is pretty much considered the top line intel chip for a laptop (but heavy due to power consumption/cooling demands)

12th Generation Intel® Core™ i9-12900HX Processor (E-cores up to 3.60 GHz P-cores up to 5.00 GHz)

But, it seems base cpu is only 2.4 (maybe 2.3). I suppose I could overclock it, but never did that before. Or maybe the turbo will always kick in when needed.

I guess I’ll see how it goes and report back! Hah!


1 Like

It is indeed their top-range. It seems Intel is going to use the same approach for all laptop cpus, because it’s the same for gen 12 I7 cpus, etc.

But my cpu has also only 2.5GHz base freq, but sometimes runs prolonged times on 3 or 4GHz, so the automatic adjustment of the cpu seems to work.

My 2 cents.

For a given TDP, when they increase the number of cores, because of the heat, they have to reduce the base frequency. Which is not an advantage if you are mainly using one core…

Thx, David, so a good reason to possibly overclock if you are not using many cores.

In this case (Lenovo 16 Thinkpad), they sort of seem to anticipate that people using few cores will overclock. I think in their marketing materials they acknowledge that this is not locked.

With my current Dell XPS 13 I have not run into a problem with CPU. I tend not to use heavy synthesis or effects. Sample libraries are more my thing.

In general, CPU use should not change much over years of use. In contrast, as I continually add sample libraries and new rackspaces to my computer, loaded ram (not using predictive loading) will continue to (gradually) increase. I would like to avoid predictive loading (though it is a great resource). I really like the luxury of immediate access to all “songs” (rackspaces).

But, I am learning now that based on default status (not overclocked) status, my current Dell XPS laptop seems to have a base cpu of 2.6 instead of 2.4 (or 2.3, I’ve read both) on my new Lenovo.

So, if necessary, I may want to consider conservative overclocking.

We’ll see…

Frank, that is good to know (I responded to David’s post first).

So, maybe their turbo works well enough that I will not need to consider overclocking.


My impression is Intel hardly mentions base frequency in new chips (I know from a marketing perspective that makes sense).

They seem to view it as a floor for low cpu tasks (idling, reading a document, etc.)

They seems say the important CPU speed to consider is the Turbo speed and that that CPU speed will automatically increase up to the turbo max whenever needed.

So, if it operates the way they say it does, maybe for their new chips base frequency is no longer too relevant (?):

See: What Is Intel® Turbo Boost Technology? - Intel

Yes but the problem is that for audio applications we don’t want to work at low frequencies, but processors have difficulties to continuously work at high turbo boost frequencies and goes to the base frequency following rules which are perhaps OK for word processing, but not for real time audio applications.

Thanks, David, yes I started reading some prior threads about the issue.

I suppose there is a possibility that this Turbo 3.0 technology will play better with audio applications (?) (Maybe not).

If not (and I have an issue), I can learn about overclocking.

Thanks again for sharing you knowledge.


I would be scared about overclocking a laptop. As overclocking means force an higher frequency and thus an higher temperature, while it can be difficult to cool a laptop significantly more. If the CPU temperature goes too high, the CPU will throttle in order to protect himself against too high temperatures. The throttle mode consist in drastically reducing the frequency for cooling down the CPU and that’s not what we want for real time audio processing.

1 Like

Fingers crossed it works okay out of the box.

The thing is, as Frank said, it seems like Intel is setting a low frequency base CPU for all its top line chips.

They seem to anticipate that they will actually operate higher than base frequency (Turbo Boost Max 3.0)

But, on the other hand, live audio performance is not their first priority…

The clock speed (commonly referred to as the frequency) of a CPU is how many instructions per second it can process. Turbo Boost is available on many Intel CPUs and is essentially a temporary overclock that increases the CPU’s frequency when additional processing power is needed. This can only happen as long as it the CPU is below a certain power, current and temperature threshold so it is not a full-time performance boost. The amount of frequency increase will vary depending on the number of cores that are in use. When fewer cores are in use, the clock speed of the remaining cores can be increased more than when all of the cores are in use.

The problem is that audio DSP processing and the software architecture of the applications, like VSTs, are quite specific in the way they operate and how they are written to use the available resources. In general, audio processing is linear and not able to use multicore, hyperthreading resources (and I am vastly oversimplifying, no flames please!). Quite a lot has been written about this here in the Community so I will let you do the research. What I usually find useful are tailored internet searches to see what various forums and manufacturers of software that use audio processing thinks works best for the current year. You get a sense of the circle of good cpus, vendors, machines and combinations and from that you should be able to make an educated guess on your particular choice. My personal guess is your choice will work just fine for what you want to do, and since you have purchased it we will all be looking for your experience with it - quite helpful for future members just like you that would like to know what others are using. How you set up GP has an significant effect too (like predictive loading and limiting how many rackspaces can load), so don’t forget to include your configuration and how you manage your gigs!

1 Like

That was great. Thank you.
I suspect I will be fine.
If there is a bottleneck, I could also explore multiple instances of GP (using another one of those extra high performance cores).


You can explore this blog article :slight_smile:[blog] Clever ways to optimize your plugin usage

Don’t worry if there is a bottleneck, I believe that Deskew is the only company that has all covered: Windows optimization, Mac optimization, plugins/best practices/tips/hacks, so don’t worry, you are at the right place :slight_smile:

(I also monitor and index tips across this community, don’t want to lose any of community gems!)


Yesterday I’ve been toying around with using a lot of plug-ins, pushing cpu (according to gp) as high as possible without distorting sound. The test:

  • Sample rate 48.000 at 64 samples
  • ThrottleStop active
  • Unfiltered Audios’ Triad to act like a heavy plug-in. This is no-way intended to make Triad look bad, I love Triad for the creative things you can do with it
  • Tenth generation I7 cpu (6/12 cores/threads), base 2.5GHz, turbo 4.5GHz
  • Focusrite Scarlett 4i4 Gen 3 interface
  • Windows 11 latest patch.

Here some empirical findings (with all the usual disclaimers):

  • Without ThrottleStop, stable sound is not possible with 64 samples anyway. 192 samples is my normal safe-go-to-haven
  • Intels’ turbo does a good job, together with ThrottleStop. When I throw in multiple Triad-instances, the cpu-frequency goes to somewhere between 4-4.5GHz. and stays there
  • Cpu in GP compared to cpu in taskmanager gives different readings, but that is explainable because GP only considers the audio-thread (for good reason)
  • Somewhere between 65%-75% cpu in GP, sound becomes unstable. Taskmanager shows that one of the cores clips at 100%. Increasing the sample buffer to 192 samples helps, probably because it lowers (also) the cpu-usage
  • Triad eats a lot of cpu of the audio thread, even when disconnected (hence the reason I used it for this test). I got at 70% cpu when using approximately 11 instances.

My main conclusion is that you do not have to make the base frequency the main consideration, you can take the turbo into account as well.

This are my cents 3 and 4 :grinning:


Thanks, Frank, that is very encouraging. (I am currently okay with a 256 sample buffer on my Dell XPS 13 with 32 GB ram. More than that bothers me. I do not have Throttle Stop installed).

I think you hit one of the core (no pun intended) questions.

Does “turbo” (not really a great name for this, in my view) allow flexible adjustment of (single core) CPU speed in real time.

David seems to take a bit more skeptical view (disclaimer, everyone here knows more than I do). And I have read posts where people complain about artifacts that seem to be due to “turbo” kicking in,

But, I think I have read more posts that the base frequency is pretty much a floor and turbo does a decent job raising the speed up in typical audio situations (up to the Turbo Boost max or less depending on other demands, heat etc.). It does not seem to be limited to very short amounts of time. There is a range between the base frequency and the top turbo frequency where (I am hoping) you can expect to operate when using GP with “typical” (whatever that is) usage.

Maybe they have improved turbo? (Now “Turbo Max 3.0”).

Well, once I get up and running (may take a bit), I will report back.

(My most important factor me in selecting the new machine (Lenova Thinkpad 16 w/ 64 GB installed ram, expandable up to 128 GB) was ram capacity so I can keep adding rackpaces years into the future without the need to create separate gig files or use predictive loading. Fingers crossed.).

It is flexible. Lets say it is “cpu on demand”, but practically it means, once you need more, you get that and it will stick there for a while (some kind of hysteresis), so al least in my situation, it doesn’t bite.

I can relate to that: Trust is nice, having control is better…

But after all, we are in the situation that we will have to put up with it, because cpus with a fixed frequency, at least with Intel, are not available anymore. The only thing you can do is switch off everything (SpeedStep, SpeedShift, Turbo, C1, etc.), but then only the base frequency remains and that is also not really what I want…

Keep us posted how your laptop performs. I’m rather eager to see. Within a few years I’ll have to buy a new laptop myself (again)…