Hello! It’s spring! February is usually completely dead in terms of music technology product and this year it hasn’t disappointed – it’s very quiet. Something about the time between the NAMM show in January and the Musikmesse at the end of March. So there’s been bugger all to talk about. So instead, and in celebration of the YouTube channel about to hit 1 million views – wayhey – I thought I’d have a bit of a rant – it’s cathartic, healthy I think to give voice to the horror so we can move on with a spring in our step. You know me – ever the optimist – because, well, everything is awesome.
So please excuse the fury and the language and rest assured that my normal cheery self will be restored very soon – hopefully by the end. So, mum if that’s you watching you might want to turn off now.
Text continues below the video:
So, Thunderbolt – what an ill conceived bunch of w**k. Laptops – just f**k off. MIDI – what a gloriously helpful and fabulous pile of inadequate s***e. Intel – what the f**k are you on about – no one understands, no one.
But first. Last month I talked about the new version of Softubes Console 1 – and I was completely wrong about how the UAD plug-in support worked.
When they said you can now run UAD plug-ins on the Console 1 – they meant that the Console 1 software could now host UAD plug-ins, not that the plug-ins could be run on the Console 1 hardware. Confused? Yes, me too. So I’d assumed that the Console 1 has a DSP chip that could run the plug-ins – oh no – you still need to have UAD hardware. So ultimately rather than running a UAD compressor as a VST plug-in you can run it inside the Console 1 software plug-in giving you slightly easier hardware control. It seems that the truth is a whole lot less exciting than the fantasy. Oh well, I guess that’s still useful – sorry if I misled anyone.
What a load of over-promised and underachieving codswallop. It’s supposed to be the perfect interconnect that removes the need to other ports and lets you do everything you like down one cable. Is it heck as like? You never know what it’s doing and what it’s going to work with. Just because it worked one way once doesn’t mean it’ll work the same next time. I did a video all about it a couple of months back just after UAD, Focusrite and Microsoft all chimed in to say how fabulous it all is. Except that both UAD and Focusrite are on TB2 and Microsoft were talking about TB3. Then we find that UAD say they don’t support TB3 if it comes on a card – which is almost every high-end system running the X99 chipset. Focusrite then say they don’t support TB3 at all – which is the one MS fixed.
I decided to make a video about it, showing the differences in performance and ways of connecting a Clarett 2Pre to one of our high-end systems. But it was so random, and flaky and inconsistent that it would be difficult to put forward something that would always work. That’s not to say it doesn’t work – it does work, it just needs a lot of fussing about for the technology to reveal itself. There tends to be two choices with high-end desktop systems in terms of motherboard – Asus and Gigabyte. Asus have gone down the road of an expansion board that you plug into a free PCIe slot and connects to a header on the board. Gigabyte did that to start with but have now integrated it on some board.
With the Asus I am constantly frustrated by the fact TB doesn’t exist in the system until you plug something into it – why the hell not? Every other device in a system has an entry in the Device Manager, but not Thunderbolt, oh no, it’s too clever for that. So you never really know if it’s installed or working until it is – which is really hard to troubleshoot. The slot the expansion card goes into has to be very specific and correlate to a setting in the BIOS – but because of the lack of obvious device manager entries it makes it hard to know whether you’ve got that right until you’ve tried every combination of slot and setting. On the Gigabyte board with integrated thunderbolt the installation is much much smoother – it’s just there and you don’t have to worry about getting the settings right. But I’ve had more blue screens and booting trouble with the Gigabyte than I’ve ever had with the Asus.
But it’s just inconsistent – the way something works on one install doesn’t work the same in another. With some things hot plugging is the only way to get it to jar itself into life and with others you have to turn everything on in a specific order for it to be found in windows.
Then don’t get me started on adapters. I had loads of trouble with the Startech one that everyone recommends – but no trouble with the Apple one. But why – it’s doing the same thing? It is flippin’ bonkers.
It’s like Firewire all over again where there was no knowing if your audio interface would work with it or not. And it’s all very well you Apple users looking smug with your new Powerbook – yes that’s great, but your Macs haven’t been updated for years so you’re not going to see TB3 anytime soon on anything powerful.
The good news is that it can and does absolutely work, it can give you fabulously low latency – and all my systems that offer Thunderbolt will leave us working perfectly – it’s just that the journey to that point can be a bit frustrating. Why can’t it be simple? Why does the Focusrite stuff and Universal stuff seem to work differently.
I think UAD and Focusrite are all really waiting for the next generation of high-end boards to arrive when it will probably all be sorted out and built in. That would be super.
Laptops laptops laptops. Why is it so difficult to find a laptop that works well for music production? This has always been the case with PC laptops. They are often full of crappy cheap components that cause all sorts of DPC latency problems when trying to do low latency audio. Badly written drivers hold onto the CPU longer than they need to and that can result in clicks and pops in your audio. But of course laptops weren’t built for music production, they were built for web browsing, watching movies, using office and playing games – none of that requires real-time audio processing. The trend in mobile technology is to put power conservation first – this means that the CPU is always trying to power down – a disaster for glitch free music production. But, you know it’s something that a lot of us want to do.
There’s no way of knowing whether any given laptop will be capable of glitch free music making – you cannot assume that it will. Two years ago when I bought my Surface Pro 3 I took a risk on it working for audio – because no one had tried it and no one had shared that information. Luckily it worked! Fabulous! Then last year I bought the Surface Pro 4 much more confidently and it was completely screwed – it wouldn’t work properly – no reason or explanation. Now, thankfully, after many updates it does work well for music – but there’s no magic formula to that. After my review of the Surface Book – which is excellent by the way – I got a ton of people asking if this Dell would be good, or this Asus or this Yoga thing and I have absolutely no idea. I cannot say that any computer will work for audio production until I have tried it myself. You cannot guarantee it. There may be something in there that’s just going to mess with things. The new Surface Studio – will that work? I don’t know! I mean this applies actually to any PC, laptop, desktop, all-in-one – doesn’t matter.
Now of course you can buy a computer from anywhere you like and it will probably work for music production. You can pick up a laptop from Tesco and perform live in front of a stadium crowd – it might work. But there are no guarantees. And if you find one that works by the time you’ve told your mates and they’ve bought one it’s a different version and might have something else that causes clicks and pops. It’s a flippin’ nightmare!
We used to do three different laptop models, they all did the job, but over the last 6 months all three have changed, been updated by the manufacturer and now they all cause glitching in the audio – we can’t supply them anymore.
Now what I should say is that there’s a difference between what I consider acceptable performance and your usage. I imagine any old laptop is going to work to some degree and many many people never push them very far and have years of trouble free music making. But when I sell a laptop as a platform for music production it has to be able to consistently put power to the road on huge projects with completely stable playback – it’s a big ask and I expect a lot from a product I’m putting my name on.
So what are you going to do? For desktops it’s easy – get one from an Audio PC supplier, from me – they will be fantastic. For laptops I don’t know – get a Surface Book or a MacBook. I can’t advise you based on a laptops specs or features Because they mean nothing until I’ve actually tried it.
Which brings me onto Intel. What the f**k is going on with your naming conventions. The number of times people ask me “I’ve got an old system with an i5 – if I get a laptop with an i7 will it be more powerful?” I have no idea – it completely depends on what sort of i5 and i7 we are talking about. Is it a desktop i5, a K series, an M series or an Ultrabook series and is it last generation, the generation before that or from 10 years ago? Is this generation i5 faster than last generation i7? Is it a Kaby lake, sky ranch, Harribo, forest gump, locked, unlocked or extreme? Is 8 cores faster than 6 cores if it’s going slower? And what about hyperthreading or virtual cores or multi-threaded thermal nuclear cores?
Does anybody know? Do I know? I don’t know if I know and I’m sure no clarity is about to be achieved any time soon.
I don’t want to give a lecture on Intel nomenclature because it’s very boring. But my advice is that when comparing Intel processors you can only really do it within the same series. Generally speaking there are 4 main series’ of processors:
U series – these are for Ultrabooks and designed to be low power, cool and quiet – fabulous battery life, modest performance. This is all the posh thin laptops like Surface and Dell Ultrabooks.
M Series – Better performing mobile processors (usually), but lower battery life, hotter, noisier
K series or desktop series – fabulous performance, desktop processor up to 4 cores for Z170 or Z270 chipset motherboards
E series – Extreme processor – top performance, high-end desktop up to 10 cores with X99 chipset.
And it’s not that simple because they are not always referred to in these terms. With the desktop processors there are versions with a “K” for unlocked and versions without and some with a “T” for being lower power. It’s almost impossible to navigate.
And then you have to think about generation. Currently we are up to 7th generation desktop and mobile processors, also called KabyLake for no suitably explored reason. With the high-end processors they are called Broadwell-E which is actually still 5th generation, but still out-performs the 7th generation desktop.
And so because of Moores Law every new generation is twice as fast as the predecessor? Well no, Moore is a bit of a con artist. The law actually refers to the number of transistors and that’s been more or less true – but it has not resulted in a doubling of performance. We seem to see a 10-20% increase in performance in real terms with each generation – but this varies.
So, calling something an i7 processor means fuck all without knowing which series and which generation, how fast and how many cores. So ever time you see an Intel advert that goes – bing bing bing bing i7 – it’s telling you nothing.
Rule of thumb – new generation beats old generation. Desktop beats mobile. More cores is better, faster speed is better and if you’re worried about whether a 4GHz Quad core is better than a 3.2GHz 6 Core then just total it up – the 4 core is giving you 16GHz, your 6 core is giving you 19.2GHz – it’s as good as comparison as anything else.
Does MIDI suck? No not really, it’s bewilderingly useful and has been for 35 years. However, it’s time for it to make way for something better. A revolution in computer technology has grown up around it – we can do phenominal things with our virtual and hardware studios and yet MIDI stays dumb, passive and surprisingly low resolution. I mean when it arrived it was awesome. At the time, the early 80’s, we had control voltage which does one thing down one cable. MIDI came along and could do a 128 things, on each of 16 channels down one cable – wow. But there it has stayed. There’s a great article I came across from Sound On Sound April 1986 where Chris Jordan talked about everything that’s wrong with MIDI and how it needs to progress in the next couple of years – none of what he suggested as happened.
So where’s the problem? Well there can be timing issues and sync issues once you are dealing with multiple devices, multiple chained interfaces and stuff – but most people don’t encounter them. It’s not really about that it’s about how there are limitations that we really should be past by now. MIDI is a serial protocol, meaning that the data is dealt with one at a time – that’s where the timing issues come when dealing with large amounts of data because it’s slow. It’s also unidirectional – it only goes one way. If I had a dollar for every time someone had plugged their MIDI cables in the wrong way round – I wouldn’t have stayed in tech support very long. So MIDI can’t ask a question and get a response – it can only send a load of data or receive a load of data on another cable. It’s also designed for keyboard players and so when guitarists and other instrument players try to get involved, using pitch tracking, MIDI pickups or alternative controllers it never works quite as well as you’d like .
There are a few recent technologies that have brought this home to me.
OSC – The first time I used TouchOSC on the iPad I was blown away by the elegance and simplicity of it. Once you’d worked out how to setup the MIDI over LAN and got your in/outs right then you turn it on and marvel as it pulled in all this data from Ableton Live and presented it to you. MIDI can’t do that. This is nicely highlighted in Yeco – a touch controller for Ableton. It has a lovely page of MIDI controls that sit there waiting for you to manually map them to things – but they are dumb, they tell you nothing. Whereas when you look at the OSC based page it’s alive with information and detail on the ever changing situation in Ableton. This is what it should all be like.
The next is MPE – multidimensional Polyphony expression. This is what ROLI keyboards and the Linnstrument use to give expressive controls to individual notes. With MIDI the controller data sent on a channel effects every note on that channel. Play a chord, push the pitchbend and it effects all the notes – that’s pretty rubbish really. With MPE you can pitch bend notes individually – or use other expressions and parameter changes on them. This is accomplished by a hack that pushes the individually notes onto different MIDI channels so they can be treated individually – but it requires hardware and software support to do it. This is something Chris Jordan called Mode 4 back in 1983 – he considered it a vital improvement to MIDI and here we are nearly 35 years later and it’s only just happening.
Ableton Link is the next one – so deliciously simple. Any device running Ableton Link on a network, wired or wireless can dictate and follow tempo. One device changes tempo and the rest follow – so simple, so easy it’s fabulous. With MIDI you have to have a master who sends out to all the other slaves – in one direction – you can’t change that situtation without stopping and recabling.
Finally we have the 128 values that forms the resolution of MIDI. 128 values isn’t bad and to most of us it sounds continuous. But to many it’s just not quiet natural enough – it doesn’t feel authentic when dealing with filters and modulations. MIDI triumphed over control voltage because it could do so much – but CV is making a big comeback these days. Partly I think because people are rediscovering how good it feels – and actually doing one thing brilliantly with one cable has a lot of merit – and perhaps we don’t really do tons of stuff over MIDI. And the other reason is that it holds a continuously varying voltage of infinite value. And that creates a feel and a sound that 7-bit MIDI just can’t match.
So what about MIDI 2.0 or “HD MIDI”? It’s been talked about for so long that I’m not sure anyone really believes it will happen. The original MIDI spec seemed to happen at a particular time that was ready for it. It was free, it was cheap, it was open to all manufacturers to implement it. Anything like that seems so unlikely these days as patents and technology are so closely guarded. Something has to be adopted by a large manufacturer to push it into existence. I think the reason why it’s becoming an issue again is that hardware is regaining its dominance. After a decade or two everything being inside the box it’s all spilling out again. But there’s no sense of it all going external – we want both hardware and software running in harmony. We want everything to be able to talk to each other seamlessly and creatively. I don’t believe MIDI is really up to that task anymore. Maybe we’ll have to cope with a hotch potch of MIDI and OSC and MPE and MIDI-to-CV, DAW to DC Coupled interface and networked sync. Or maybe someone could come up with a protocol that’s do it all with no cables and no setup.
Phew, so that’s a load off my chest – I feel great now.
So is anything coming up? Yes – Bitwig released version 2 of Bitwig Studio and I’ll be doing a full review of that very soon. And also my journey into Modular synthesis is taking shape – I’ve bought a case and I’m currently poised to buy my first modules – so that’s very exciting. Check it out in this channel or on MoltenModular.com
If you like my videos then please share, subscribe and the recommend the channel to people and let’s see if I can crack that 1 million views. At the moment I’m on 993,884.
In the meantime – go and make some tunes.