Chkcpu wrote on 2025-01-19, 19:25:

Hi Cyanopsis, […]
Show full quote

Hi Cyanopsis,

About having only a very short time to press DEL to enter the BIOS Setup, I haven’t experience this before.
Did you disable the full screen QDI bootup logo in the BIOS FEATURES SETUP menu? If so, you should have ample time to hit the DEL key, all through the memory test and the Drive scan on the IDE channels.
Also setting the HARD DISK TYPE and MODE to AUTO for all 4 drives in the STANDARD CMOS SETUP menu will give you more time.

It's a bit difficult to see a pattern, but if I spam DEL on POST, it usually reboots. If I press DEL on memory count once or twice, it boots to BIOS.

I have been thoroughly analyzing the SpeedEasy BIOS code, especially the Bootblock part where SpeedEasy does the actual programm […]
Show full quote

I have been thoroughly analyzing the SpeedEasy BIOS code, especially the Bootblock part where SpeedEasy does the actual programming of the RC5051 Vcore voltage controller on this board.
If my analysis is correct, it is possible that with a K6-2+/III+ installed, the initial Vcore is set to 2.7V before the SpeedEasy control in the main BIOS POST module kicks in. However, this only happens with a K6-2+ and K6-III+ CPU and with the CPU Voltage Ctrl set to AUTO in the SpeedEasy CPU Setup menu. With the CPU Voltage Ctrl set to Manual and the Core Voltage set to 2.0V or 2.1V, this initial 2.7V will not happen and the Bootblock will set the selected Manual Voltage directly.

This means that when you have the Core Voltage set to Manual 2.0V, and you still experience the coldstart boot problem on the 1.6V K6-2+, the above high initial Vcore is not the issue and the boot problem may be in the CPU itself.

I assume that the CMOS battery is good and that you don’t see a “CMOS checksum Error” or “A new CPU is installed” message. If you do, then you also have a startup Vcore of 2.7V because the BIOS has lost its CMOS data and doesn’t know what CPU is installed yet.

I will install the 1.6v CPU and to some more testing (another k6-2+ is currently installed). I'll get back to you. Thank you!

Chkcpu wrote on 2025-01-19, 19:25:

This means that when you have the Core Voltage set to Manual 2.0V, and you still experience the coldstart boot problem on the 1.6V K6-2+, the above high initial Vcore is not the issue and the boot problem may be in the CPU itself.

I assume that the CMOS battery is good and that you don’t see a “CMOS checksum Error” or “A new CPU is installed” message. If you do, then you also have a startup Vcore of 2.7V because the BIOS has lost its CMOS data and doesn’t know what CPU is installed yet.

The computer doesn't even boot when switching to AUTO for core voltage. I have to clear CMOS for it to come back to life. I'm guessing the BIOS is booting in a safety mode when it detects a new CPU, so it's able to boot once with vcore set to AUTO. But if I leave the setting as is and save, there's no POST after reboot. So I'm pretty sure I've always had vcore set to MANUAL, as per your instructions for the BIOS mod. Not sure I want to risk anything by trying AUTO with my 2.0V K6-2+ CPU (that always boots normally) just for comparison, but maybe the AUTO 2.2V is too much for the 1.6V CPU to handle?

Hi Cyanopsis,

Thanks for your feedback.
In the meantime I’ve made a new patch J.2 BIOS version for your QDI P5MVP3/A3 that addresses the possible high initial Vcore condition on the K6-2+/III+. I’ve send you a Personal Message with this new BIOS. Please use this J.2 BIOS for further testing.

It is good that you used the SpeedEasy CPU Voltage control on Manual for the K6-2+/III+. This setting avoids the high initial Vcore condition from happening. 😉

So the most important change in the J.2 BIOS is a fix for the condition that could cause a 2.7V Vcore value after installing a K6-2+/III+ while leaving the SpeedEasy CPU Voltage control at Auto. This bugfix prevents this and should always limit the Auto Vcore on the K6-2+/III+ to 2.2V. This 2.2V is the maximum specified Vcore for the K6-2+ so it shouldn’t harm the 1.6V model or prevent it from booting.
Of course it is still advisable to set the required Vcore for the K6-2+ manually at 2.0V via Jumper Emulation.

While I was working on this BIOS, I’ve added 2 more improvements to the SpeedEasy menu:
- The K6-2CXT, K6-III, and K6-2+/III+ CPU models remap the x2 multiplier to x6 internally. I’ve changed the x2 multiplier indication in Jumper Emulation mode to show x6.0 on these CPUs, to reflect the actual multiplier.
- I’ve added deviating multiplier support for the WinChip 2A CPU. The proper X2.5, X3, X3.33, and X3.5 multipliers are now indicated.
For all other functions, the patch J.2 BIOS is identical to the J.1 version from 2018.
Let us know how the J.2 BIOS works and if it solves the 1.6V K6-2+ coldboot issue.

Cheers, Jan

Chkcpu wrote on 2024-10-23, 18:53:

Curiously, I never had a HOT-566 BIOS patch request before, until now. That’s why it’s not on my Unofficial K6plus webpage yet. […]
Show full quote

0xDEADBEEF wrote on 2024-10-23, 00:26:

I was wondering if it is possible to patch Shuttle HOT-566 (https://theretroweb.com/motherboards/s/shuttle-hot-566) for faster AMD K6 CPUs. I am running latest BIOS with K6-2 450 (with 75Mhz FSB) but the BIOS reports incorrect CPU frequency and I wonder whether it could be affecting other stuff, since sometimes I get an NMI. I tried different memory and it didnt help so I think it is not coming from memory, but maybe BIOS programs something wrong when it doesn't detect CPU frequency correctly. I see other shuttle BIOSes on the unofficial page but not HOT-566 ;(

Curiously, I never had a HOT-566 BIOS patch request before, until now.
That’s why it’s not on my Unofficial K6plus webpage yet. 😉

Yes, I can patch the latest 566WAQ0D BIOS for better K6-2(+)/III(+) support and fix the 32GB IDE HDD limit bugs as well. However, I have several other BIOS projects to finish first so it may take a while…

Looking into this 566WAQ0D BIOS, I found that the x6 multiplier isn’t supported for any FSB. That’s why you get strange speed readings when running your K6-2 at 6x75. This may also cause the busspeed calculation in the BIOS to fail, resulting in incorrect programming of the chipset timing registers for memory and L2 cache. So this setting may indeed cause instabilities.

You can try to lower the multiplier to x5.5 for a 412MHz CPU speed. This 5.5x75 setting is supported by the BIOS and should run stable if the instability was caused by the not supported x6 multiplier. In fact, all Multiplier/FSB combinations of x2.5, x3.0, x3.5, x4.0, x4.5, x5.0, x5.5 and FSB 50, 60, 66, 75, and 83MHz are supported, except for the x5.5/83 setting.

If you still experience an NMI, even at a 5.5x66 setting, then the BIOS is probably not the cause.
Do you use an ISA SoundBlaster soundcard? If so, remove this card and see if that fixes the NMI errors. Depending on the configuration of these SB cards, they can produce these NMI errors.

Greetings, Jan

Hi 0xDEADBEEF,

The patched Shuttle HOT-566 BIOS is ready, and I send you a PM with this BIOS so you can test it. Please let us know how it works.

Apart from added K6-2+/III+ support, I’ve also expanded the CPU speed support up to 500MHz and fixed the 32GB and 64GB IDE HDD limit bugs and the UDMA bug.

I welcome other HOT-566 owners to test this patched 566WAQ0D BIOS. Just send me a PM if you like to join the patched BIOS test team.

I friendly warning:
Don’t run a K6-2+/III+ CPU with the original BIOS on this HOT-566 board when using the Auto Vcore function. The original 566WAQ0D BIOS puts a very unhealthy 3.2V Vcore on a K6-2+/III+ in Auto mode, instead of the required 2.0V!
So use the manual Vcore settings via jumpers on the board when trying to run a K6plus with the original BIOS.

In the patched BIOS, I expanded the Auto Vcore logic to include the K6plus.
Just after power-up, the initial start Vcore in Auto on a K6-2(+)/III(+) should now be 2.2V. This initial Vcore setting is done very early during POST by the Bootblock code, so that these K6-2(+)/III(+) CPUs are protected against a high Vcore when the BIOS hangs early due to a corrupted BIOS or bad or missing RAM.
During normal POST the final Vcore setting quicky follows when the BIOS reaches POST step 31. This POST_31 routine does the detailed Vcore setting and should set 2.2V for the K6 model 7/K6-2, 2.4V for the K6-III, and 2.0V for the K6-2+/III+.

Cheers, Jan

So is there a brief time where there is 3.3v going to the cpu? Or is the VRM idle while the boot block executes?

@Chkcpu

I'm working on this board;

https://theretroweb.com/motherboards/s/tmc-my … t-ver.-1.1#bios

It has the option to select SCSI in the boot order menu in the BIOS but it never initialises my Promise SATA 150 TX4 BIOS during boot time.

The interface is present and working - once in a Windows 98 environment I can see the contents of drives attached to it. I just can't boot from it...

Do you think this is something that could be modded to fix? Is it worth heading down this rabbit hole? The BIOS versions that are available are all circa 1997 so I imagine there are quite a few things that could be fixed/modded/improved

Keen to know your thoughts before I investigate much further...

I think I usually set boot to C and it boots my sata try a few options.

Sphere478 wrote on 2025-02-15, 05:47:

I think I usually set boot to C and it boots my sata try a few options.

Do you see your SATA controller's BIOS initialise after post? I see this on all my slot 1 boards etc but not this socket 7. Even with Boot drive set to something else, I'd still expect to see the sata controller BIOS screen before the boot drive routine kicks in

Sphere478 wrote on 2025-02-14, 17:50:

So is there a brief time where there is 3.3v going to the cpu? Or is the VRM idle while the boot block executes?

Hi Sphere478,

When I started looking at a number of jumperless socket 7 BIOSes, I was also intrigued how these boards control the initial CPU voltage before the reset line goes low and the CPU starts executing code.
Without jumpers or dipswitches to set the voltage for the CPU, and with the wide range of required voltages for various S7 CPUs, you need the motherboard to provide a sufficient high voltage for 2.8V/2.9V Vcore CPUs to start reliably but not too high for 2.1V/2.2V models to get damaged.

I found that these jumperless S7 boards usually start at 2.6V Vcore for Dual voltage CPUs and 3.4V for Single voltage parts, a nice compromise. Single/Dual voltage switching is usually done in hardware with the help of the CPU’s Vcc2DET# pin.
When the CPU starts executing code, the BIOS will take over and adjusts this value for the installed CPU. Often this is just a coarse setting, 2.2V for low Vcore CPUs and 2.8V for high Vcore models.
This initial Vcore control by the BIOS is done by the Bootblock code, during its first POST step (POST_C0 for Award BIOS). This is even before the first RAM test so the CPU is protected from a too high voltage when the BIOS hangs at a later POST step due to bad or missing RAM.

Much later during POST, the BIOS will fine-tune the voltage further for the exact CPU model. This can be an “Auto” value or a manual CPU voltage setting done earlier via the BIOS Setup menu’s.

The above means that patching these jumperless BIOSes for new CPU models is quite challenging , because both the Bootblock and main BIOS POST code needs to be analyzed and adapted if necessary. 😉

Cheers, Jan

Thanks for the details.

So ideally we would just use manual jumper settings to avoid that momentary spike.

2.6v for a fraction of a second probably isn’t the end of the world for a + cpu. So the bios fix is probably also acceptable.

Good work Jan!

I assume there is no third option that could be implemented that could get these boards to default to 2.2 on hardware level but also let bios keep control after boot block?

Sphere478 wrote on 2025-02-16, 04:03:

I assume there is no third option that could be implemented that could get these boards to default to 2.2 on hardware level but also let bios keep control after boot block?

Now you mention it, actually there IS a third option to get an automatic low Vcore on hardware level.

Apart from the Vcc2DET# pin (AL-1), which is pulled low by all socket 7 split voltage CPUs (except the non-modded Intel Tillamook) to signal a Dual-voltage CPU, there is a second signal pin on certain CPU models.
Reading the datasheets of 2.0V-2.4V Vcore socket 7 CPUs, I noticed that these CPUs have an additional Vcc2H/L# pin (AN-5) to signal a low Vcore model. On 2.8V or higher Vcore CPU models (and the Tilly), this pin is INC.

On most socket 7 boards, where the Vcore is directly controlled by jumpers, this Vcc2H/L# output from the CPU is not used. On jumperless boards however, this signal would be an ideal candidate for control of the VRM controller’s VID3 line, to lower the Vcore by 0.8V!

I know of only one S7 board where this Vcc2H/L# signal is actually used: the Freetech P5F85/87. When selecting 2.8V Vcore is the BIOS Setup, you actually get 2.0V on a K6-2+/III+. Quite scary but it works!
But if there is one board that uses the Vcc2H/L# signal, there must be others. 😉

Cheers, Jan

Lostdotfish wrote on 2025-02-14, 22:49:

@Chkcpu […]
Show full quote

@Chkcpu

I'm working on this board;

https://theretroweb.com/motherboards/s/tmc-my … t-ver.-1.1#bios

It has the option to select SCSI in the boot order menu in the BIOS but it never initialises my Promise SATA 150 TX4 BIOS during boot time.

The interface is present and working - once in a Windows 98 environment I can see the contents of drives attached to it. I just can't boot from it...

Do you think this is something that could be modded to fix? Is it worth heading down this rabbit hole? The BIOS versions that are available are all circa 1997 so I imagine there are quite a few things that could be fixed/modded/improved

Keen to know your thoughts before I investigate much further...

Hi Lostdotfish,

That TI5TT is a nice full-sized ATX Socket 7 i430TX board. With its wide range of Vcore settings and 75MHz FSB support, it is ideal for running a K6-2+ at 450MHz. 😉

Yes, there are indeed a lot of things that can be fixed in its 1997 BIOS to improve CPU and IDE HDD support.
I can do that for you, however I have no ready fix for the Promise SATA 150 TX4 boot issue. So I have to search the BIOS POST routines for the initialization logic of option ROMs.
Not seeing the SATA controller’s BIOS initialization message is a strong clue that the main BIOS is missing a step here.

So let me know if you like a patched BIOS for this board and I will put it on my list.
Btw, is it the Y6D version you are running now?

Greetings, Jan

Chkcpu wrote on 2025-02-16, 17:39:

Hi Lostdotfish, […]
Show full quote

Lostdotfish wrote on 2025-02-14, 22:49:

@Chkcpu […]
Show full quote

@Chkcpu

I'm working on this board;

https://theretroweb.com/motherboards/s/tmc-my … t-ver.-1.1#bios

It has the option to select SCSI in the boot order menu in the BIOS but it never initialises my Promise SATA 150 TX4 BIOS during boot time.

The interface is present and working - once in a Windows 98 environment I can see the contents of drives attached to it. I just can't boot from it...

Do you think this is something that could be modded to fix? Is it worth heading down this rabbit hole? The BIOS versions that are available are all circa 1997 so I imagine there are quite a few things that could be fixed/modded/improved

Keen to know your thoughts before I investigate much further...

Hi Lostdotfish,

That TI5TT is a nice full-sized ATX Socket 7 i430TX board. With its wide range of Vcore settings and 75MHz FSB support, it is ideal for running a K6-2+ at 450MHz. 😉

Yes, there are indeed a lot of things that can be fixed in its 1997 BIOS to improve CPU and IDE HDD support.
I can do that for you, however I have no ready fix for the Promise SATA 150 TX4 boot issue. So I have to search the BIOS POST routines for the initialization logic of option ROMs.
Not seeing the SATA controller’s BIOS initialization message is a strong clue that the main BIOS is missing a step here.

So let me know if you like a patched BIOS for this board and I will put it on my list.
Btw, is it the Y6D version you are running now?

Greetings, Jan

Hey Jan,

Thanks for your reply. I've just double checked this board (it's the first time I've had it out for about 4-5 years). I think back when I added the board photos to https://theretroweb.com/ there were no BIOS images available - I've just updated to the Y6D and the board now inserts the SATA controller option ROM during boot!

I need to do some more testing as the last time I had the board out it would not post with my K6-2 300 (AFR) or my Cyrix 6x86 MX PR200. Just waiting on a socket 7 cooler to arrive as I can't find my box of them anywhere (typical...)

Either way - yes please to a modded BIOS when you have time. Happy to test and I have a epromm programmer for recovery in case of a bad flash.

EDIT

Curiosity got the better of me - found a spare chipset heatsink and fan to use as a temporary cooling solution. The board now posts fine with both the Chompers K6-2 300 and the Cyrix MX PR200

An interesting point to note - the motherboard only has 4 dip switches for speed selection, with the other 4 switch positions being unpopulated... I think I will order an 8 switch block and solder it in to see if I can get it to set 75MHz FSB and 4.5 multiplier (currently I can only run the K6-2 300 at 233 MHz

This may have just become my favourite retro motherboard...

EDIT 2

What K6 CPU should I look for to test with? That K6-2 300 AFR is the only one I have on hand but I can probably source something else. K6-2+ 450?

Oh btw, I forgot to say, some boards to boot you need to use specific pci slots. Also try that.

Chkcpu wrote on 2025-02-16, 11:11:

Now you mention it, actually there IS a third option to get an automatic low Vcore on hardware level. […]
Show full quote

Sphere478 wrote on 2025-02-16, 04:03:

I assume there is no third option that could be implemented that could get these boards to default to 2.2 on hardware level but also let bios keep control after boot block?

Now you mention it, actually there IS a third option to get an automatic low Vcore on hardware level.

Apart from the Vcc2DET# pin (AL-1), which is pulled low by all socket 7 split voltage CPUs (except the non-modded Intel Tillamook) to signal a Dual-voltage CPU, there is a second signal pin on certain CPU models.
Reading the datasheets of 2.0V-2.4V Vcore socket 7 CPUs, I noticed that these CPUs have an additional Vcc2H/L# pin (AN-5) to signal a low Vcore model. On 2.8V or higher Vcore CPU models (and the Tilly), this pin is INC.

On most socket 7 boards, where the Vcore is directly controlled by jumpers, this Vcc2H/L# output from the CPU is not used. On jumperless boards however, this signal would be an ideal candidate for control of the VRM controller’s VID3 line, to lower the Vcore by 0.8V!

I know of only one S7 board where this Vcc2H/L# signal is actually used: the Freetech P5F85/87. When selecting 2.8V Vcore is the BIOS Setup, you actually get 2.0V on a K6-2+/III+. Quite scary but it works!
But if there is one board that uses the Vcc2H/L# signal, there must be others. 😉

Cheers, Jan

Great info!

I need to get myself one of these automatic boards for my collection.

Which would you say is the best, pros cons between them?

Hello @Chkcpu !

I recently start restoring a nice MB 8500TVX (-A) rev 2.1 .
I like this small motherboard and I am planning to push it to its limits 😀.
Right now its running with a simple mod to accept pentium MMX (a simple pull-up resistor) and my 233 MMX works very well, I also fully reccaped the board.

I ordered a K6-2 400 AFR, and already prepared the mod to set down the VCore to 2.2.
Confirmed working using a scope and can easily configure different values...
But now I am waiting for my CPU to be delivered and finally testing this !

Based on other posts, it should work, but without the CPU being recognized properly by the BIOS.

Would you be able to patch it ? 😀
And... if you want / can also look for other stuff ( K6-2+ / 3 / 3+, 128GB HDD) that would be also very nice.
Later I might also try to mod the FSB to go higher than 66MHZ (changing the PLL ) , does that requires BIOS support ?

Based on a few posts, the latest official BIOS is: TVX0619B (https://theretroweb.com/motherboard/bios/tvx0 … d5530604105.zip)
I am running it on mine, it works well but has at least the 32GB limit.

I tried to have a look at that and was able to extract the different BIOS modules using AMIBCP.exe but then I dont know where to start...
I am a C developer but I confess my Assembly skill is close to zero ...

chacha wrote on 2025-02-25, 23:26:

Hello @Chkcpu ! […]
Show full quote

Hello @Chkcpu !

I recently start restoring a nice MB 8500TVX (-A) rev 2.1 .
I like this small motherboard and I am planning to push it to its limits 😀.
Right now its running with a simple mod to accept pentium MMX (a simple pull-up resistor) and my 233 MMX works very well, I also fully reccaped the board.

I ordered a K6-2 400 AFR, and already prepared the mod to set down the VCore to 2.2.
Confirmed working using a scope and can easily configure different values...
But now I am waiting for my CPU to be delivered and finally testing this !

Based on other posts, it should work, but without the CPU being recognized properly by the BIOS.

Would you be able to patch it ? 😀
And... if you want / can also look for other stuff ( K6-2+ / 3 / 3+, 128GB HDD) that would be also very nice.
Later I might also try to mod the FSB to go higher than 66MHZ (changing the PLL ) , does that requires BIOS support ?

Based on a few posts, the latest official BIOS is: TVX0619B (https://theretroweb.com/motherboard/bios/tvx0 … d5530604105.zip)
I am running it on mine, it works well but has at least the 32GB limit.

I tried to have a look at that and was able to extract the different BIOS modules using AMIBCP.exe but then I dont know where to start...
I am a C developer but I confess my Assembly skill is close to zero ...

Hello chacha,

You may have noticed that I mainly patch Award BIOSes for Socket 7.
I did patch a few AMI BIOSes for K6-2+/III+ support, but for this AMI patch to work the BIOS already needs to have support for the regular K6-2/III.

Your nice MB-8500TVX is a bit older and even the latest TVX0619B BIOS reportedly doesn’t work with an IDT WinChip or the original AMD K6. Unfortunately I have no solution for that.

However, reading the reply from @Repo Man11 in another 8500TVX thread gave me an idea.
AMD K6 in a Socket 7 8500TVX motherboard
He tells about a BIOS upgrade he got from Evergreen for use with their Upgrade CPU on an 8500TVX. He reports it worked fine with a WinChip and a K6-2/450 at 400MHz!

So I searched for this BIOS and I believe I found it. 😀
It is a 03/16/2000-i430VX-2A59GB0BC-00 BIOS update for your board from Unicore Software that came with the Evergreen Spectra K6-2/400 upgrade CPU package. This 03/16/2000 BIOS has the 32GB and 64GB HDD limit bugs fixed for full 128GiB HDD support!
In addition, it supports all socket 7 CPUs from Intel, AMD, Cyrix, Winchip, and Rise. So also the K6-2+/K6-III+ are supported.

Note that this is an Award BIOS, so you may have to use Uniflash to replace the AMI BIOS.

I’ve tested this BIOS in the Emulated Biostar MB-8500TVX-A machine of 86Box and it works great. Hopefully you have equal success on the real hardware. 😉
Please let us know how this Unicore BIOS works.

Cheers, Jan

Thank you very much Jan !

I am currently in a business trip so cannot test it yet, but as soon as I ll be back, be sure I will 😀
I should have receive the CPU by then , so I will tell you all if it works.

Cheers

PS: Yes, I did notice you were more familiar with Award BIOSes 😀

Some good news !

So, lets start:
- BIOS flash ran perfectly, using uniFlash, and it boots without issues then (and faster + bios has more options)
( I haven't tried bigger HDD yet)
- I finalized my undervolt motdto have exactly 2.2V, and while I was away I received my K6-IIE 400 (AFR).
- Installed, and with BF0/1 out => it boots at 233MHz ! Nice start !
- Then I tried to set multiplier to 2x because I am expecting this CPU to actually run at 6x if 2x set... no POST 😒
- I tried to pull-up in case it was the same problem as with the MMX CPU, no change.
- other configuration does work (underclock even more)
- Then I did the BF2 mod => short to ground
- TADA ! --==> CPU now runs at 5.5x 66 | 366MHz <===--

I can see a very significant improvment in general performances, thats nice !
But as you probably can guess I'd like it to run at 400MHz 😀 .

But I cannot figure out why it wont run at normal speed.
I tried other jumper configuration ,with BF2 shorted, and they all does work, so there is no pull up/ down issue.

I thought it can be a VRM issue ... but 366MHz seems verry stable, I would expect 400MHz to at least POST.
And as its a linear voltage regulator, lowering the voltage might actually increase power dissipation (not sure how K6 regulate tho..). I added a bigger capacitor at the transistor output and it did not change anything.
I ordered a more recent replacement Transistor, in theory it should accept 4 more amps than this one.
I am tempted to try to feed this transistor with the PSU 3.3V instead of MB 5V to lower the voltage difference...

Or maybe it has nothing to do with the hardware and this CPU just does not has the 2x => 6x feature ?
I will receive the K6-III+ 400ATZ, I hope that one will work.
I expect it to be stronger, in particular because it should be able to cache the whole RAM, but also it should require less current.

If anyone has idea on how to achieve 400MHz... I am ready for trials 😀
I dont think it can be BIOS related at that stage isn't it ?

I have received modified BIOS for Shuttle HOT-566 via PM, sadly cannot respond because I am a noob here. So thank you so much, it works great.