|
Avid Pro Audio CommunityHow to Join & Post • Community Terms of Use • Help Us Help YouKnowledge Base Search • Community Search • Learn & Support |
|
|
Thread Tools | Search this Thread | Display Modes |
#21
|
|||
|
|||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
kroon's 2009 IS a cheesegrater
|
#22
|
||||
|
||||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
Ah...did say pretrashcan. All I saw was trashcan.
|
#23
|
||||
|
||||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
Makes sense as to why those processors can't handle low buffers.
|
#24
|
|||
|
|||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
Quote:
I eventually had to up the buffer to 256 for mixdown, but now *that's* an old machine. I then upgraded to the first core 2 duo desktops, the 2.93ghz 2 core and it was a revelation. project went from showing 90% on cubase meter to 30% and i was able to add extra tracks to it at 64 buffer. Then i went to a core 2 duo 2.4ghz macbook pro in 2008, and Logic 8. All my recording was done with a firewire interface at 64 buffer. I went to an imac 3.33 i think it was, dual core, never ever went above 64 buffer with my motu firewire ultralite. Then to a cheesegrater hex and same thing.. The OP has said he is having the issue with a fresh project and no effects.. he gets cpu spikes of 70% for just 24 tracks of plain audio recording. it sounds to me like the SSD is taking time from the HD native card. I am going to say now, there should be absolutely NO problem recording 24 tracks of audio at 64 buffer, on a 12 core 3.33ghz cheese grater, unless the pcie bus is getting too much asked of it. I am going to give my advice, and the OP can try it or not, it's totally up to them, but I'll say what I believe. And firstly, i believe the pci ssd is somehow getting priority over the HD interface.. since the pcie ssd is what is being recorded to.. i think 2 blazingly fast pci ssd's are saturating the pcie bus, and that's exactly why higher buffers are working.. it's *only* at lower buffers that this sort of problem would manifest.. think sort of the equivalent of dpc latency issues in windows... You do NOT need these pci ssds for audio. sata SSD's are plenty fast for anything you can throw at them regarding audio. The ssd's the OP uses are only needed for ultra high res multitrack video. Now.. I also want to add, the apogee can not possibly have anything to do with it.. The actual connection to your computer is being made by the HD pcie card itself, the apogee is being used as converters interfacing with that card basically. It doesn't matter what you use, avid or third party converters, the problem will persist in this particular setup. SO.. my advice is 1) just try this.. remove the pcie SSD's and have just the HD native in there 2) I have heard of plenty of issue with ge force 680's in cheese grater macs and audio before.. note, along with the SSD's, your pci bus has a lot being demanded of it. THAT'S where the new gen of chipsets are much much better, way more bandwidth. But to not expect to be able to record 64 tracks of audio on your machine at 64 buffer is frankly ludicrous. Try and change it to the cheap fanless geforce included with the mac. I have seen them go for like $25.. Was it a 520 or something? I can't quite remember, sorry. 3) Put in a standard sata or sata ssd. Try again with PT 2018.1 and OS 12.12.6 Get back to me if it didn't work. But i am 90% sure it will fix it. Anyone who tells you sata ssd's are not fast enough for your needs is wrong. I will stake that claim right here and now. a sata 7200rpm drive is fast enough for 128 tracks of audio being recorded simultaneously. This is a fact. Therefore a sata SSD won't even blink at 24 tracks of audio, any sample rate. But you are at 48K and that's what I'm going on with my comparisons. Actually, you will need 2 sata drives, one Os one audio.. i really hope you have access to at least try some. I have tons of 7200's here, if you were close by i'd just give you one to try it, so you wouldn't have to buy 2 sata ssd's. I have spoken to a member via PM at gearslutz who has a cheesegrater six core and tracks at 64 buffer with HDN all day, 32 tracks plus. I've seen the videos with my own eyes. And i know from my own experience that I could do this in 2006 myself with XP and then Win7, and then Leopard and Snow leopard.. with "worse" processors than the OP. I am telling you, it's either the SSD's or the graphics card, or a combination of both. That's the issue, stealing time from the HDN card. PS sata ssd's are fast enough for all tasks required of a DAW.. included sample streaming, recording, and OS.
__________________
- Intel 14900K/NzXt Kraken Elite/64GB Kingston DDR5 6000 mhz (32x2)/ Asus Pro Art Z790/Asus 4090/Win 11 Pro 23H2/UAD Apollo 8 x2 w TBolt 3 card u/g/UAD Twin X. |
#25
|
|||
|
|||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
Quote:
|
#26
|
||||
|
||||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
I truly hope this thread doesn't get overly "personal."
__________________
~ tom thomas Formerly hobotom Pro Tools Ultimate 2024 HDX Hybrid HD Omni and 192 I/Os Windows 10 Intel Hexcore i7 All Samsung Pro SSDs Ampex MM1200 2" 24 trk tape Outboard: UREI, Eventide, Lexicon, Yamaha, TC Electronics, Orban, ART, EchoAudio, Dolby, Hughes, API, Neve, Audio Arts, BBE, Aphex, Berringer, MOTU, dbx, Allison, etc. Plug-ins: Too many to talk about. www.metrostudios.com |
#27
|
|||
|
|||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
so it's starting again.
I actually pressed the button to see his post, cause of your post Tom are you going to allow this targeted bullying to continue? And why doesn't he give a counter argument to sat why it's "nonsense"? (when in fact, it's correct.. the thunderbolt HD card is not being given priority on the pcie bus wth all that other stuff there). What can it hurt the OP to simply try it? did i say something evil? Or did i spend half hour of my time trying to help someone? Considering someone i know solved the same issue the same way, why is it nonsense? I will also add i'll recommend the OP check every pcie slot in the mac pro for the HD card. Now, i made a point.. i said that if these few members were not purposely trolling me, they would just mute me and stop quoting me. This is the same member that accused me of being on drugs in the now deleted dsp topic. I am being targeted and harassed, and i am going to the avid ceo about this, and i guarantee you this is going to end in a court case against the user. I'll be asking avid for their details. I do not take kindly to persistent bullying, and i'll be fighting back. I can't understand how the mods can be so blind.. if this user did not want trouble with me he would either mute me or stop quoting me. The fact that he can only quote me to be nasty, or accuse me of taking drugs behind my back and having a laugh with someone else at my expense, says everything. That is absolutely, definitive infallible proof, that he is purposely harassing. by the way i am happily recording all day at a friend's cheesegrater hex core at 64 samples in HDN, 32 inputs. So eat that, musicman. I am saying right here right now, publicly, to prove i do not want any tension here, that I will happily never talk about or to musicman again as long as i live, if he gives me the same courtesy. Proof i don't want trouble. Same goes for the other 4 i have had an issue with, who know who they are. Now remember, one of these 4 told me to look at myself as it must be my fault, but this topic is once again proof that it isn't. I did NOTHING here to deserve a troll.
__________________
- Intel 14900K/NzXt Kraken Elite/64GB Kingston DDR5 6000 mhz (32x2)/ Asus Pro Art Z790/Asus 4090/Win 11 Pro 23H2/UAD Apollo 8 x2 w TBolt 3 card u/g/UAD Twin X. |
#28
|
||||
|
||||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
This is not aimed at any single member, but...
While respectable disagreements on substance are fine here, negative personal comments, attacks, arguments, or name calling are not. I suggest that it stops now. If problems are escalated, it wont be pretty. And we don't want to see valuable members be suspended for things that do not need to be said. If emotion can't be kept at bay, please choose not to respond at all. Let's keep the DUC a peaceful place. Thank you.
__________________
~ tom thomas Formerly hobotom Pro Tools Ultimate 2024 HDX Hybrid HD Omni and 192 I/Os Windows 10 Intel Hexcore i7 All Samsung Pro SSDs Ampex MM1200 2" 24 trk tape Outboard: UREI, Eventide, Lexicon, Yamaha, TC Electronics, Orban, ART, EchoAudio, Dolby, Hughes, API, Neve, Audio Arts, BBE, Aphex, Berringer, MOTU, dbx, Allison, etc. Plug-ins: Too many to talk about. www.metrostudios.com |
#29
|
||||
|
||||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
Quote:
I've also seen where people have had better luck with the 6 core Macs over the 12 core. |
#30
|
|||
|
|||
Re: Is my MONSTER 12-core 3,33GHz Mac Pro underperforming on low buffer sizes?
Quote:
Re the hex core, as i said, indeed, i have now had hands on with a hex core that works with HDN at 64. So that correlates with your info also. Even on windows, i remember some users back when i was using cubase, had dual cpu systems, and would have issues with low latency.. Dual opterons i think it was back then. And people with singles wouldn't. So yes maybe it has something to do with it "power" wise though, it doesn't make any sense. It could very well be that a new dual cpu system of say 36 cores or something would do exactly the same thing, but it's not due to lack of power. As i said i tracked at 64 samples as far back as 2006. Prior that i had a pentium III 1K with dual pulsar cards, so i only used the pulsar's own ULL mixer for tracking through. But the fact remains that in every single mac i have had since then, i have always used 64 buffer.. even on my crappy 2008 core 2 duo.. the only time i have upped it to 128 for the first time in a decade, is because of pro tools performance with playing Vi's in realtime under 128 (in cubase or logic i'd be on 32 buffer). I am absolutely certain though that i could track 64 channels through core audio at 32 buffer as long as i wasn't using effects. His 12 core mac pro would eat my imac 4 core for breakfast.. My imac is only a sandy bridge 4 core 3.4ghz, and i HAVE tested 32 channels through the PT mixer at ULL already, by bypassing apollo console and using pro tools to monitor instead. I recorded all external synths in one pass at 32 buffer, all 32 analog ins. I will do it again tomorrow. I *can't* do it if i add even one reverb, but a couple EQ's and dynamics is OK. I almost want to buy a cheap 12 core cheese grater from ebay to find out for myself, as i really maintain there is something that is not right about this. The only person i know with a 12 core is using a hdspe rme pcie, and runs at 64 samples in Cubase all day every day. You are saying that you can't even record one at 256 without spikes.. Ok, if my previous post was totally incorrect, then i would say it has to be a problem with the way the HDN card operates in the pcie bus of a dual processor cheese grater, cause the RME works fine. Edit.. i just read sardi's post.. i agree 100%.. as he said, we were doing this with computers which were a fraction as powerful as an iphone. Also Bob, you say in sig you have a video pcie editing solution, as well as HDX. so i am a little confused.. i thought HDX was independent of buffer... Could you tell me the specs of the machine you can't record at lower than 256 buffer in? Even if it's the 2.4ghz 12 core, it still should be powerful enough to do what the OP is asking, and his is much more powerful than that. Also what video card, the 520? And whatever other pcie devices are also being used when the overload occurs. PS, as i said someone i knew had the same issue and solved it after i gave him the same advice, by using only sata SSD's (his mac pro uses all 4 available sata bays) and removing everything pcie other than a very low grade ge force video card and the HDN.. even removed his usb 3 lacie pcie card.. the HDN is absolutely the only extra thing in the pcie bay. It shouldn't have to be this way, but it fixed it for him.. However, he "only" has the hexa core 3.33ghz. His buffer is permanently on 64 samples. So it really might be a dual processor issue.
__________________
- Intel 14900K/NzXt Kraken Elite/64GB Kingston DDR5 6000 mhz (32x2)/ Asus Pro Art Z790/Asus 4090/Win 11 Pro 23H2/UAD Apollo 8 x2 w TBolt 3 card u/g/UAD Twin X. Last edited by TNM; 02-10-2018 at 11:06 AM. Reason: added PS |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
6-core (3.33GHz) or 2 quad-core (2.4GHz) Westmere | Detlef Brockmann | Pro Tools TDM Systems (Mac) | 9 | 02-21-2020 10:39 AM |
8 GB RAM in MacPro 6 core 3.33GHZ | songman | Pro Tools TDM Systems (Mac) | 1 | 12-07-2011 04:05 AM |
6-core (3.33GHz) or 2 quad-core (2.4GHz) Westmere | Detlef Brockmann | macOS | 4 | 02-18-2011 07:08 PM |
Mac Pro 3.33Ghz 6 core Westmere & VE PRO | johnpaulgeorgeringo | Pro Tools TDM Systems (Mac) | 36 | 01-02-2011 09:35 AM |
2010 mac pro 2.4Ghz 8 core vs 3.33Ghz 6 core | edadema | Pro Tools TDM Systems (Mac) | 32 | 09-07-2010 07:36 AM |