Will it be possible for the 15inch 2.2 MBP to have a Thunderbol eGPU??

[KernalPanic]

Originally Posted by Mr MM

what is laughable, is that even with the 6750m turned on, in windows the mac still provides the good battery life of 5h

Given the fact MUCH more powerful cards with no hybrid at all and no consideration for battery life at all are getting 3-4, its pretty much expected.

[Mr MM]

Originally Posted by Lieto

eGpu over TB... does it mean i wont need to sell my mbp in a year or two?
Will apple allow that? heh

Its not apple to allow, but us to code something, people on the windows world face some difficulties sometimes doing the egpu thing

Originally Posted by KernalPanic

Given the fact MUCH more powerful cards with no hybrid at all and no consideration for battery life at all are getting 3-4, its pretty much expected.

actually no, windows machines pack that same gpu, or nvidia similars, only achieve that battery life with extended batteries and switchable gpus.

its is really a shame that customizations or actual engineering like what happens on business machines is not applied to those cases, so yeah its pitful at best to have the same battery life as machines that uses some form of switching


now on the thread, I havent been able to identify what ports are wired to the thunderbolt port, however there is one good news, even my mbp 13 posses a mux in the gpu... so we may be able to wire the thunderbolt port to display the image on the notebook lcd, no external monitors required

[ryzeki]

Originally Posted by hawk1410

Thunderbolt has more bandwidth than express card slot, I think I read a post that we should get around 80% of true performance of a desktop card.

Of course it does, because Express card is barely 1x of PCIe. If it where 1 lane of PCIe v3.x, the Express card slot would have almost the same bandwidth as thunderbolt. It is far much better to get an Express card slot with actual 4x or more lanes, which would give significantly more bandwidth than thunderbolt.

Originally Posted by Karant-rex

YEah, even i read on Nando's thread that Thunderbolt should be able to give about 80% of true performance, I am happy with that. As long as it is faster than a Express card slot. I plan on using it with a GTX560 Ti or a GTX570.

10-20Gb/s is not that much, considering typical current PCIe speeds roam about 64Gb/s for the old v2.0 PCIe. v3.x is even higher. It is a lot, however, when compared to the rest of ports (USB etc) available on PC.

It's a nice port with high memory bandwidth compared to many other ports available in PC, but for graphics card, its far easier(and faster) to make a higher lane count express card for an eGPU.

[Mr MM]

Originally Posted by ryzeki

10-20Gb/s is not that much, considering typical current PCIe speeds roam about 64Gb/s for the old v2.0 PCIe. v3.x is even higher. It is a lot, however, when compared to the rest of ports (USB etc) available on PC.

It's a nice port with high memory bandwidth compared to many other ports available in PC, but for graphics card, its far easier(and faster) to make a higher lane count express card for an eGPU.

well according to some reviews, x4 is an acceptable bandwidth, most games suffer a penalty of 1-10%, being some ports the cause of high consumption of bandwidth and where staggering differences where felt, something around 50%

I did some tests right now, and nothing is conclusive as of yet, but it seems that the pcie ports that the thunderbolt is going to use are the ports 4-8, thus raising the possibility that all are sequential.

[nando4]

Originally Posted by Mr MM

I did some tests right now, and nothing is conclusive as of yet, but it seems that the pcie ports that the thunderbolt is going to use are the ports 4-8, thus raising the possibility that all are sequential.

Found a Macbook Pro 8.2's lspci output here.

What they've done is split the x16 root port into two x8 ports (0:1.0 and 0:1.1). The first has all 8 lanes connected and is used to host the dGPU (HD6750M). The second is the Thunderbolt port with only 4 of the 8 lanes connected as highlighted below. So Thunderbolt is x4 2.0 as I expected it to be.

00:01.1 PCI bridge [0604]: Intel Corporation 2nd Generation Core Processor Family PCI Express Root Port [8086:0105] (rev 09) (prog-if 00 [Normal decode])
	Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
	Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- SERR- TAbort- Reset- FastB2B-
		PriDiscTmr- SecDiscTmr- DiscTmrStat- DiscTmrSERREn-
	Capabilities: [88] Subsystem: Apple Computer Inc. Device [106b:00dc]
	Capabilities: [80] Power Management version 3
		Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold+)
		Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
	Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
		Address: feeff00c  Data: 4159
	Capabilities: [a0] Express (v2) Root Port (Slot+), MSI 00
		DevCap:	MaxPayload 128 bytes, PhantFunc 0, Latency L0s <64ns, L1 <1us
			ExtTag- RBE+ FLReset-
		DevCtl:	Report errors: Correctable- Non-Fatal- Fatal- Unsupported-
			RlxdOrd- ExtTag- PhantFunc- AuxPwr- NoSnoop-
			MaxPayload 128 bytes, MaxReadReq 128 bytes
		DevSta:	CorrErr- UncorrErr- FatalErr- UnsuppReq- AuxPwr- TransPend-
		LnkCap:	Port #3, Speed 5GT/s, Width x8, ASPM L0s L1, Latency L0 <256ns, L1 <4us
			ClockPM- Surprise- LLActRep- BwNot+
		LnkCtl:	ASPM Disabled; RCB 64 bytes Disabled+ Retrain- CommClk+
			ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
		LnkSta:	Speed 5GT/s, Width x4, TrErr- Train- SlotClk+ DLActive- BWMgmt+ ABWMgmt-
		SltCap:	AttnBtn- PwrCtrl- MRL- AttnInd- PwrInd- HotPlug- Surprise-
			Slot #2, PowerLimit 75.000W; Interlock- NoCompl+
		SltCtl:	Enable: AttnBtn- PwrFlt- MRL- PresDet- CmdCplt- HPIrq- LinkChg-
			Control: AttnInd Unknown, PwrInd Unknown, Power- Interlock-
		SltSta:	Status: AttnBtn- PowerFlt- MRL- CmdCplt- PresDet+ Interlock-
			Changed: MRL- PresDet+ LinkState-
		RootCtl: ErrCorrectable- ErrNon-Fatal- ErrFatal- PMEIntEna+ CRSVisible-
		RootCap: CRSVisible-
		RootSta: PME ReqID 0000, PMEStatus- PMEPending-
		DevCap2: Completion Timeout: Not Supported, TimeoutDis- ARIFwd-
		DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis- ARIFwd-
		LnkCtl2: Target Link Speed: 5GT/s, EnterCompliance- SpeedDis-, Selectable De-emphasis: -3.5dB
			 Transmit Margin: Normal Operating Range, EnterModifiedCompliance- ComplianceSOS-
			 Compliance De-emphasis: -6dB
		LnkSta2: Current De-emphasis Level: -3.5dB
	Capabilities: [100 v1] Virtual Channel
		Caps:	LPEVC=0 RefClk=100ns PATEntryBits=1
		Arb:	Fixed- WRR32- WRR64- WRR128-
		Ctrl:	ArbSelect=Fixed
		Status:	InProgress-
		VC0:	Caps:	PATOffset=00 MaxTimeSlots=1 RejSnoopTrans-
			Arb:	Fixed+ WRR32- WRR64- WRR128- TWRR128- WRR256-
			Ctrl:	Enable+ ID=0 ArbSelect=Fixed TC/VC=ff
			Status:	NegoPending- InProgress-
	Capabilities: [140 v1] Root Complex Link
		Desc:	PortNumber=03 ComponentID=01 EltType=Config
		Link0:	Desc:	TargetPort=00 TargetComponent=01 AssocRCRB- LinkType=MemMapped LinkValid+
			Addr:	00000000fed19000
	Kernel driver in use: pcieport
	Kernel modules: shpchp

[funky monk]

If you're worrying about bandwidth, you should probably watch these videos.